Ampere Altra Max 128 Core vs. AMD EPYC vs. Intel Xeon Linux Performance - Ubuntu 23.10 Server Benchmarks

Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2312011-NE-AMPERE21815
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

Chess Test Suite 2 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 7 Tests
CPU Massive 17 Tests
Creator Workloads 4 Tests
Cryptography 3 Tests
Database Test Suite 5 Tests
Fortran Tests 5 Tests
HPC - High Performance Computing 14 Tests
Imaging 2 Tests
Common Kernel Benchmarks 4 Tests
Linear Algebra 2 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 5 Tests
Multi-Core 21 Tests
OpenMPI Tests 11 Tests
Programmer / Developer System Benchmarks 7 Tests
Python Tests 4 Tests
Scientific Computing 7 Tests
Server 7 Tests
Server CPU Tests 12 Tests
Common Workstation Benchmarks 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 9554
November 13 2023
  9 Hours, 25 Minutes
EPYC 9654
November 15 2023
  9 Hours, 20 Minutes
EPYC 9684X
November 08 2023
  11 Hours, 12 Minutes
EPYC 9754
November 09 2023
  10 Hours, 41 Minutes
Xeon Platinum 8490H
October 31 2023
  8 Hours, 28 Minutes
Ampere Altra Max 128c
November 27 2023
  13 Hours, 36 Minutes
Invert Hiding All Results Option
  10 Hours, 27 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

Xeon Platinum 8490H

Processor: Intel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads), Motherboard: Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS), Chipset: Intel Device 1bce, Memory: 512GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9684X

Processor: AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9754

Changed Processor to AMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads).

Processor Change: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116

EPYC 9554

Changed Processor to AMD EPYC 9554 64-Core @ 3.10GHz (64 Cores / 128 Threads).

EPYC 9654

Changed Processor to AMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads).

Ampere Altra Max 128c

Processor: ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores), Motherboard: GIGABYTE MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS), Chipset: Ampere Computing LLC Altra PCI Root Complex A, Memory: 512GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Intel I350

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Ampere Altra Max 128 Core vs. AMD EPYC vs. Intel Xeon Linux Performance - Ubuntu 23.10 Server BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsNetworkMonitorOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionXeon Platinum 8490HEPYC 9684XEPYC 9754EPYC 9554EPYC 9654Ampere Altra Max 128cIntel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads)Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS)Intel Device 1bce512GB3201GB Micron_7450_MTFDKCC3T2TFSASPEEDUbuntu 23.106.6.0-rc5-phx-patched (x86_64)GNOME Shell 45.0X Server 1.21.1.7GCC 13.2.0ext41920x1200AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads)AMD Titanite_4G (RTI1007B BIOS)AMD Device 14a4768GBBroadcom NetXtreme BCM5720 PCIeAMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads)AMD EPYC 9554 64-Core @ 3.10GHz (64 Cores / 128 Threads)AMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads)ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores)GIGABYTE MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS)Ampere Computing LLC Altra PCI Root Complex A512GBVGA HDMI2 x Intel I3506.6.0-rc5-phx-patched (aarch64)1920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- Xeon Platinum 8490H: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9684X: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9754: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9554: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - EPYC 9654: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Ampere Altra Max 128c: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v Processor Details- Xeon Platinum 8490H: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1- EPYC 9684X: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9754: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116- EPYC 9554: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- EPYC 9654: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e- Ampere Altra Max 128c: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Java Details- OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Details- Python 3.11.6Security Details- Xeon Platinum 8490H: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9684X: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9754: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9554: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - EPYC 9654: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Ampere Altra Max 128c: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Xeon Platinum 8490HEPYC 9684XEPYC 9754EPYC 9554EPYC 9654Ampere Altra Max 128cLogarithmic Result OverviewPhoronix Test SuiteeasyWaveNAS Parallel BenchmarksminiBUDELiquid-DSPlibxsmmSysbenchoneDNNJohn The RipperXmrigOpenSSLACES DGEMMPrimesieveStockfishTimed Godot Game Engine CompilationTimed Node.js CompilationAlgebraic Multi-Grid BenchmarkasmFishminiFEClickHouseApache CassandraXcompact3d Incompact3dCoremarkRocksDBTimed LLVM CompilationASKAP7-Zip CompressionNWChemHelsingTimed Linux Kernel CompilationRodiniaStress-NGCloverLeafMemcachedPostgreSQLLULESHDuckDBHigh Performance Conjugate GradientTimed Gem5 CompilationGraphicsMagickRawTherapeeGraph500

Xeon Platinum 8490HEPYC 9684XEPYC 9754EPYC 9554EPYC 9654Ampere Altra Max 128cLogarithmic Per Watt Result OverviewPhoronix Test SuiteXmrigSysbenchNAS Parallel Benchmarks7-Zip CompressionlibxsmmCoremarkApache CassandraGraphicsMagickminiBUDEasmFishRocksDBLiquid-DSPminiFEStockfishACES DGEMMMemcachedAlgebraic Multi-Grid BenchmarkStress-NGClickHouseJohn The RipperASKAPOpenSSLHigh Performance Conjugate GradientGraph500LULESHP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.MP.W.G.M

Ampere Altra Max 128 Core vs. AMD EPYC vs. Intel Xeon Linux Performance - Ubuntu 23.10 Server Benchmarksnpb: LU.Cjohn-the-ripper: WPA PSKnpb: BT.Ceasywave: e2Asean Grid + BengkuluSept2007 Source - 2400easywave: e2Asean Grid + BengkuluSept2007 Source - 1200npb: IS.Djohn-the-ripper: MD5xmrig: Wownero - 1Mliquid-dsp: 240 - 256 - 512minibude: OpenMP - BM2minibude: OpenMP - BM2openssl: ChaCha20-Poly1305openssl: ChaCha20npb: FT.Cliquid-dsp: 128 - 256 - 512libxsmm: 256npb: CG.Caskap: tConvolve MT - Griddingsysbench: CPUlibxsmm: 128openssl: AES-256-GCMincompact3d: input.i3d 193 Cells Per Directionnpb: MG.Conednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUopenssl: AES-128-GCMaskap: tConvolve MPI - Griddingstress-ng: Vector Floating Pointopenssl: SHA256compress-7zip: Decompression Ratingstress-ng: Vector Mathstress-ng: AVX-512 VNNImt-dgemm: Sustained Floating-Point Rateprimesieve: 1e13openssl: SHA512build-linux-kernel: defconfigbuild-godot: Time To Compilerocksdb: Rand Readclickhouse: 100M Rows Hits Dataset, Third Runbuild-nodejs: Time To Compilestress-ng: Matrix Mathamg: asmfish: 1024 Hash Memory, 26 Depthminife: Smalljohn-the-ripper: bcryptjohn-the-ripper: Blowfishclickhouse: 100M Rows Hits Dataset, Second Runcassandra: Writesclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachepgbench: 100 - 1000 - Read Onlypgbench: 100 - 1000 - Read Only - Average Latencycoremark: CoreMark Size 666 - Iterations Per Secondbuild-llvm: Ninjaxmrig: Monero - 1Maskap: tConvolve MPI - Degriddingnpb: SP.Cstress-ng: Wide Vector Mathmemcached: 1:100graphics-magick: Sharpennwchem: C240 Buckyballstress-ng: Memory Copyingrocksdb: Read While Writingcompress-7zip: Compression Ratinghelsing: 14 digitstress-ng: Fused Multiply-Addstress-ng: Floating Pointgraph500: 26incompact3d: X3D-benchmarking input.i3drodinia: OpenMP LavaMDduckdb: TPC-H Parquetlulesh: hpcg: 144 144 144 - 60graph500: 26stress-ng: NUMAduckdb: IMDBbuild-gem5: Time To Compilerawtherapee: Total Benchmark Timegraph500: 26graph500: 26memcached: 1:10build-linux-kernel: allmodconfigpgbench: 100 - 1000 - Read Write - Average Latencypgbench: 100 - 1000 - Read Writegraphics-magick: Enhancedstockfish: Total Timecloverleaf: clover_bm64_shortcloverleaf: clover_bm16Xeon Platinum 8490HEPYC 9684XEPYC 9754EPYC 9554EPYC 9654Ampere Altra Max 128c136342.91399141180757.43118.55445.8502885.71957733335870.210137666672146.28885.85122321658475030981011082770989.979059800001030.135560.258112.07154451.201764.664924196254012.798342088049.90447.50576864412176022365.2138052.1962987799233299902315559.624947898.6623.25738852.3752244582044735.602128.467268374486487.01151.567304345.30161182600013081859435452.59387793913485.91139654461.9017322130.5772233945.606722169.52827608.519616.072703.354423715.194380269.266912056.618809.55609111838969086.758145542113.6622454.70333129000353.55759542.713148.09923997.71531.24324544830001144.0499.249197.75946.578111312000010738600004667128.01289.96614.16470604112415346926538.93324.17333543.81611601312135.7360.25224.5796138.541479966773567.313381666674251.197170.048345798034863494325321817124391.0011423000003147.859557.0814731.5388317.803102.47697782883207.30814419137608.68684.11689675013479767965.5214129.85124596517757623489585746.568459429.3840.32006225.9694010563958036.59198.555486862785558.78119.203417406.39237485333323392663152937.6166119166080546.14286460534.2535259450.2833703841.717094131.99069205.658310.1208538.322597371.597930429.137751641.626495.63951845255441164.349202387450.7829220.78383754000440.54858430.317147.48924712.88423.77845012220001173.07116.050180.06152.9918821810008540290005928614.58283.86514.79267606127728198271151.31298.29279880.91779799296367.0472.35529.5885773.251445666777607.217147000005892.337235.694461374746590657652605747141255.0713845333333343.544431.5812660.8514483.582702.510096344497309.02806168127404.161037.89117424463972327269.2280779.02165326815057774122773130.2311411104.7443.68192321.7565364807820341.545118.247611259161445.46130.460552067.04229104966729026050550393.7204828204677431.65236320425.5134611570.2894543138.116767146.38629356.126593.4131257.153437154.967669596.619241700.834497.511104604558657148.947256387572.7239031.64377033000493.50245225.150177.13422356.74625.89185021510001397.61147.601208.57666.132118451000011470900005552637.52319.42613.98971483145136332964249.03459.16253193.84512773233622.4456.88322.4525184.361317933360032.411760000003883.576155.343287552025123403275660767124827.7511452333332524.645867.4410327.5311423.412400.46493858378179.74551621125921.29470.72975365181576337265.9181377.97107534517597521569507522.167240174.3630.17490428.1363362202853033.516103.500376983415571.53121.928346518.02232115600021016705450351.4138035138032546.67301963535.5535156230.2843056532.141324131.20347926.431177.2119472.302250265.278108606.406691452.122827.20741265048973671.268167642337.9024168.36368016000515.79275834.014143.30624037.44222.4511467453000967.56103.720179.74349.9438651520008361590006291456.53238.28913.72672860114224177432845.82460.24271216.04649345261864.1159.53523.5075195.241576300075127.414002000004400.504176.020356731270590510715410203125164.0311721000002858.854465.0412332.4454951.332578.67810247405138.45212936118608.07665.04091041693177347003.2215564.78130355422273648630611736.598271540.0639.28215824.1644045932361036.600101.563429310263554.30120.156421053.02229673066725020902050320.2166850166887551.97285324531.5236885790.2713884296.768928133.07059047.741081.6125373.872597630.058412361.968091568.428310.82970411956495061.529202941117.4530824.97391167000430.08032930.240146.27623701.44232.12834993470001237.36115.805180.85752.0518399170008161720006136092.71284.24912.78678212131928906690947.98429.4639811.9611986353396.60326.368122.5501196.3331430001909.23711233331353.36954.13511226053281316177637030035153.27366046667917.217272.734295.72437954.92940.830649222021023.885224042291.841452.303828502741002416.54102559.01101167218107541464399080.054692150.5418.11499441.6753443496079778.547227.783435437276253.26268.704682356.13106065600021243247924122.0108753108827253.53227548250.3021232170.4722829772.917301266.5434322.15285.3620218.432355041.964860402.431274269927163.45879191132837856.895151397996.3922187.72224032000606.52170831.791237.16116218.33121.23993332390001454.68143.781260.32666.0459877700009804630005131209.29307.87816.75859693123018820206362.77550.30OpenBenchmarking.org

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70K140K210K280K350KSE +/- 114.80, N = 3SE +/- 686.23, N = 5SE +/- 1817.56, N = 5SE +/- 1940.05, N = 15SE +/- 1937.17, N = 6SE +/- 157.58, N = 439811.96271216.04253193.84279880.91333543.81136342.911. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H400800120016002000206.801373.591351.511461.211835.75500.46
MinAvgMaxAmpere Altra Max 128c23.1192.5222.8EPYC 965425.4197.5305.2EPYC 955422.2187.3292.6EPYC 975416.3191.5315.7EPYC 9684X29.4181.7285.8Xeon Platinum 8490H58.6272.4332.7OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 11037 Joules per run.2. EPYC 9654: Approximate power consumption of 2527 Joules per run.3. EPYC 9554: Approximate power consumption of 2435 Joules per run.4. EPYC 9754: Approximate power consumption of 2592 Joules per run.5. EPYC 9684X: Approximate power consumption of 2059 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 5449 Joules per run.

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200K400K600K800K1000KSE +/- 13.33, N = 3SE +/- 64.33, N = 3SE +/- 23.00, N = 3SE +/- 2344.03, N = 3SE +/- 68.67, N = 3SE +/- 184.59, N = 3119863649345512773779799611601399141-m64-m64-m64-m64-m641. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H5001000150020002500832.362050.561683.292240.722055.661319.84
MinAvgMaxAmpere Altra Max 128c23.2144.0169.5EPYC 965427.4316.7362.0EPYC 955422.7304.6347.4EPYC 975479.8348.0391.3EPYC 9684X30.6297.5341.8Xeon Platinum 8490H59.0302.4358.5OpenBenchmarking.orgWatts, Fewer Is BetterJohn The Ripper 2023.03.14CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4992 Joules per run.2. EPYC 9654: Approximate power consumption of 10872 Joules per run.3. EPYC 9554: Approximate power consumption of 10357 Joules per run.4. EPYC 9754: Approximate power consumption of 11832 Joules per run.5. EPYC 9684X: Approximate power consumption of 10116 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10383 Joules per run.

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70K140K210K280K350KSE +/- 130.49, N = 3SE +/- 1485.73, N = 4SE +/- 560.21, N = 4SE +/- 174.81, N = 4SE +/- 1185.12, N = 5SE +/- 44.76, N = 353396.60261864.11233622.44296367.04312135.73180757.431. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500276.371161.121083.181404.071445.11624.61
MinAvgMaxAmpere Altra Max 128c23.3193.2217.5EPYC 965426.5225.5320.2EPYC 955421.8215.7302.5EPYC 975417.2211.1316.4EPYC 9684X29.4216.0305.6Xeon Platinum 8490H60.1289.4374.1OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 11528 Joules per run.2. EPYC 9654: Approximate power consumption of 3721 Joules per run.3. EPYC 9554: Approximate power consumption of 3721 Joules per run.4. EPYC 9754: Approximate power consumption of 3377 Joules per run.5. EPYC 9684X: Approximate power consumption of 3110 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 6077 Joules per run.

easyWave

The easyWave software allows simulating tsunami generation and propagation in the context of early warning systems. EasyWave supports making use of OpenMP for CPU multi-threading and there are also GPU ports available but not currently incorporated as part of this test profile. The easyWave tsunami generation software is run with one of the example/reference input files for measuring the CPU execution time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereasyWave r34Input: e2Asean Grid + BengkuluSept2007 Source - Time: 2400Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350SE +/- 1.24, N = 3SE +/- 0.40, N = 3SE +/- 0.33, N = 3SE +/- 0.45, N = 3SE +/- 0.14, N = 3SE +/- 0.36, N = 3326.3759.5456.8872.3660.25118.551. (CXX) g++ options: -O3 -fopenmp
MinAvgMaxAmpere Altra Max 128c22.782.6101.2EPYC 965426.8165.7196.4EPYC 955422.1141.7162.3EPYC 975478.3141.4159.5EPYC 9684X30.2175.5209.0Xeon Platinum 8490H69.5217.9249.7OpenBenchmarking.orgWatts, Fewer Is BettereasyWave r34CPU Power Consumption Monitor601201802403001. Ampere Altra Max 128c: Approximate power consumption of 27623 Joules per run.2. EPYC 9654: Approximate power consumption of 10881 Joules per run.3. EPYC 9554: Approximate power consumption of 8975 Joules per run.4. EPYC 9754: Approximate power consumption of 11216 Joules per run.5. EPYC 9684X: Approximate power consumption of 11758 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 27159 Joules per run.

OpenBenchmarking.orgSeconds, Fewer Is BettereasyWave r34Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H306090120150SE +/- 1.12, N = 3SE +/- 0.14, N = 3SE +/- 0.06, N = 3SE +/- 0.20, N = 15SE +/- 0.32, N = 3SE +/- 0.04, N = 3122.5523.5122.4529.5924.5845.851. (CXX) g++ options: -O3 -fopenmp
MinAvgMaxAmpere Altra Max 128c23.179.0104.0EPYC 965427.4155.2203.4EPYC 955420.9131.5161.8EPYC 975474.5136.9162.3EPYC 9684X29.6165.4209.2Xeon Platinum 8490H59.4205.9240.1OpenBenchmarking.orgWatts, Fewer Is BettereasyWave r34CPU Power Consumption Monitor601201802403001. Ampere Altra Max 128c: Approximate power consumption of 10299 Joules per run.2. EPYC 9654: Approximate power consumption of 4551 Joules per run.3. EPYC 9554: Approximate power consumption of 3681 Joules per run.4. EPYC 9754: Approximate power consumption of 4856 Joules per run.5. EPYC 9684X: Approximate power consumption of 5018 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10707 Joules per run.

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H13002600390052006500SE +/- 0.34, N = 3SE +/- 13.89, N = 5SE +/- 14.93, N = 5SE +/- 18.26, N = 5SE +/- 65.80, N = 5SE +/- 6.63, N = 41196.335195.245184.365773.256138.542885.711. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H9182736458.92830.16732.75939.26436.17011.604
MinAvgMaxAmpere Altra Max 128c23.9134.0177.6EPYC 965425.5172.2306.4EPYC 955421.7158.3289.5EPYC 975418.3147.0271.0EPYC 9684X30.2169.7293.9Xeon Platinum 8490H64.4248.7349.1OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4422 Joules per run.2. EPYC 9654: Approximate power consumption of 2377 Joules per run.3. EPYC 9554: Approximate power consumption of 2216 Joules per run.4. EPYC 9754: Approximate power consumption of 1970 Joules per run.5. EPYC 9684X: Approximate power consumption of 2274 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 5160 Joules per run.

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H3M6M9M12M15MSE +/- 3785.94, N = 3SE +/- 3214.55, N = 3SE +/- 3179.80, N = 3SE +/- 835422.69, N = 15SE +/- 55480.73, N = 3SE +/- 10493.38, N = 33143000157630001317933314456667147996679577333-m64-m64-m64-m64-m641. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: MD5Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H11K22K33K44K55K18999.5051091.3942662.2252624.0449767.4428959.28
MinAvgMaxAmpere Altra Max 128c24.3165.4181.9EPYC 965427.1308.5358.3EPYC 955422.1308.9356.5EPYC 975478.5274.7358.3EPYC 9684X29.1297.4338.6Xeon Platinum 8490H67.9330.7377.0OpenBenchmarking.orgWatts, Fewer Is BetterJohn The Ripper 2023.03.14CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 10642 Joules per run.2. EPYC 9654: Approximate power consumption of 19848 Joules per run.3. EPYC 9554: Approximate power consumption of 19874 Joules per run.4. EPYC 9754: Approximate power consumption of 17747 Joules per run.5. EPYC 9684X: Approximate power consumption of 19131 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 21166 Joules per run.

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1MAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H17K34K51K68K85KSE +/- 4.67, N = 3SE +/- 60.06, N = 4SE +/- 18.19, N = 3SE +/- 604.78, N = 4SE +/- 35.19, N = 4SE +/- 13.06, N = 31909.275127.460032.477607.273567.335870.2-maes-maes-maes-maes-maes1. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s Per Watt, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1MAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H6012018024030018.61270.21212.04284.00265.76114.36
OpenBenchmarking.orgWatts, Fewer Is BetterXmrig 6.18.1CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 23.62 / Avg: 102.62 / Max: 112.56Min: 26.88 / Avg: 278.03 / Max: 364.83Min: 22.57 / Avg: 283.12 / Max: 354.21Min: 79.72 / Avg: 273.26 / Max: 349.95Min: 30.36 / Avg: 276.81 / Max: 356.76Min: 59.68 / Avg: 313.66 / Max: 376.621. Ampere Altra Max 128c: Approximate power consumption of 54489 Joules per run.2. EPYC 9654: Approximate power consumption of 5074 Joules per run.3. EPYC 9554: Approximate power consumption of 6040 Joules per run.4. EPYC 9754: Approximate power consumption of 4850 Joules per run.5. EPYC 9684X: Approximate power consumption of 5052 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10142 Joules per run.

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 240 - Buffer Length: 256 - Filter Length: 512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H400M800M1200M1600M2000MSE +/- 115518.16, N = 3SE +/- 1044030.65, N = 3SE +/- 723417.81, N = 3SE +/- 3752776.75, N = 3SE +/- 2195702.87, N = 3SE +/- 1311911.24, N = 3371123333140020000011760000001714700000133816666710137666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 1.6Threads: 240 - Buffer Length: 256 - Filter Length: 512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1.4M2.8M4.2M5.6M7M2036261.915116900.214404001.146362747.945019125.783222808.58
MinAvgMaxAmpere Altra Max 128c22.9182.3217.3EPYC 965426.9273.6312.6EPYC 955420.0267.0306.7EPYC 975480.2269.5303.1EPYC 9684X30.4266.6306.2Xeon Platinum 8490H67.4314.6377.3OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 1.6CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 6440 Joules per run.2. EPYC 9654: Approximate power consumption of 9395 Joules per run.3. EPYC 9554: Approximate power consumption of 9257 Joules per run.4. EPYC 9754: Approximate power consumption of 9342 Joules per run.5. EPYC 9684X: Approximate power consumption of 9154 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 11010 Joules per run.

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H13002600390052006500SE +/- 0.16, N = 3SE +/- 13.67, N = 3SE +/- 34.21, N = 3SE +/- 1.57, N = 3SE +/- 0.19, N = 3SE +/- 11.04, N = 31353.374400.503883.585892.344251.202146.29-mcpu=native-march=native-march=native-march=native-march=native-march=native1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H50100150200250SE +/- 0.01, N = 3SE +/- 0.55, N = 3SE +/- 1.37, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.44, N = 354.14176.02155.34235.69170.0585.85-mcpu=native-march=native-march=native-march=native-march=native-march=native1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -lm
OpenBenchmarking.orgBillion Interactions/s Per Watt, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H0.19670.39340.59010.78680.98350.3060.6190.5460.8740.6340.271
MinAvgMaxAmpere Altra Max 128c23.6177.0186.4EPYC 965427.0284.6328.3EPYC 955421.6284.4320.8EPYC 975418.3269.8323.2EPYC 9684X30.0268.1318.0Xeon Platinum 8490H59.5316.4375.5OpenBenchmarking.orgWatts, Fewer Is BetterminiBUDE 20210901CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 17584 Joules per run.2. EPYC 9654: Approximate power consumption of 9296 Joules per run.3. EPYC 9554: Approximate power consumption of 10523 Joules per run.4. EPYC 9754: Approximate power consumption of 7015 Joules per run.5. EPYC 9684X: Approximate power consumption of 9206 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 20249 Joules per run.

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20-Poly1305Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H100000M200000M300000M400000M500000MSE +/- 1451328.25, N = 3SE +/- 174121043.89, N = 3SE +/- 5365828.17, N = 3SE +/- 165042020.15, N = 3SE +/- 183485436.83, N = 3SE +/- 242044451.24, N = 3112260532813356731270590287552025123461374746590345798034863223216584750-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: ChaCha20-Poly1305Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H300M600M900M1200M1500M617930879.771089518475.95915099689.881276157031.331068328648.36657459512.26
MinAvgMaxAmpere Altra Max 128c23.7181.7199.4EPYC 965427.9327.4352.7EPYC 955423.0314.2350.5EPYC 975481.1361.5380.6EPYC 9684X29.8323.7342.0Xeon Platinum 8490H75.0339.5352.9OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 33428 Joules per run.2. EPYC 9654: Approximate power consumption of 59918 Joules per run.3. EPYC 9554: Approximate power consumption of 57609 Joules per run.4. EPYC 9754: Approximate power consumption of 66161 Joules per run.5. EPYC 9684X: Approximate power consumption of 59342 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 62357 Joules per run.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H140000M280000M420000M560000M700000MSE +/- 5187436.26, N = 3SE +/- 31096600.10, N = 3SE +/- 10027126.50, N = 3SE +/- 75341939.23, N = 3SE +/- 12204219.35, N = 3SE +/- 249978665.97, N = 3161776370300510715410203403275660767657652605747494325321817309810110827-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: ChaCha20Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H400M800M1200M1600M2000M851511981.271550153775.141260085671.451795931120.231518907337.02956989711.80
MinAvgMaxAmpere Altra Max 128c23.3190.0213.6EPYC 965427.6329.5356.9EPYC 955422.7320.0353.5EPYC 975480.8366.2393.9EPYC 9684X31.4325.4345.4Xeon Platinum 8490H67.5323.7377.1OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1102203304405501. Ampere Altra Max 128c: Approximate power consumption of 34958 Joules per run.2. EPYC 9654: Approximate power consumption of 60401 Joules per run.3. EPYC 9554: Approximate power consumption of 58567 Joules per run.4. EPYC 9754: Approximate power consumption of 67013 Joules per run.5. EPYC 9684X: Approximate power consumption of 59557 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 59567 Joules per run.

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30K60K90K120K150KSE +/- 41.38, N = 4SE +/- 698.81, N = 8SE +/- 1162.69, N = 15SE +/- 1190.42, N = 8SE +/- 686.49, N = 8SE +/- 445.75, N = 635153.27125164.03124827.75141255.07124391.0070989.971. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2004006008001000244.94837.59904.171081.10812.37306.03
MinAvgMaxAmpere Altra Max 128c26.5143.5215.4EPYC 965426.1149.4304.5EPYC 955422.3138.1302.0EPYC 975418.4130.7305.6EPYC 9684X30.2153.1294.6Xeon Platinum 8490H55.2232.0359.8OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 2619 Joules per run.2. EPYC 9654: Approximate power consumption of 1308 Joules per run.3. EPYC 9554: Approximate power consumption of 1178 Joules per run.4. EPYC 9754: Approximate power consumption of 1241 Joules per run.5. EPYC 9684X: Approximate power consumption of 1340 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 2590 Joules per run.

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H300M600M900M1200M1500MSE +/- 16666.67, N = 3SE +/- 556776.44, N = 3SE +/- 1233333.33, N = 3SE +/- 1026861.45, N = 3SE +/- 2458319.21, N = 3SE +/- 1999408.25, N = 336604666711721000001145233333138453333311423000009059800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s Per Watt, More Is BetterLiquid-DSP 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1.1M2.2M3.3M4.4M5.5M2009607.204496781.754256153.305240555.274519511.452858526.50
MinAvgMaxAmpere Altra Max 128c23.0182.1216.6EPYC 965427.0260.7297.0EPYC 955422.6269.1306.3EPYC 975478.9264.2293.6EPYC 9684X30.9252.7286.4Xeon Platinum 8490H71.8316.9376.1OpenBenchmarking.orgWatts, Fewer Is BetterLiquid-DSP 1.6CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 6314 Joules per run.2. EPYC 9654: Approximate power consumption of 9036 Joules per run.3. EPYC 9554: Approximate power consumption of 9238 Joules per run.4. EPYC 9754: Approximate power consumption of 9071 Joules per run.5. EPYC 9684X: Approximate power consumption of 8762 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10882 Joules per run.

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 256Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H7001400210028003500SE +/- 11.05, N = 3SE +/- 36.77, N = 3SE +/- 5.95, N = 3SE +/- 3.30, N = 3SE +/- 2.46, N = 3SE +/- 2.55, N = 3917.22858.82524.63343.53147.81030.1-march=armv8.1-a-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.21. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 256Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H481216206.77011.02610.39917.81212.2023.026
MinAvgMaxAmpere Altra Max 128c23.3135.5211.9EPYC 965425.9259.3357.9EPYC 955422.3242.8346.6EPYC 975418.7187.7336.6EPYC 9684X29.9258.0354.9Xeon Platinum 8490H56.4340.4356.9OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 37664 Joules per run.2. EPYC 9654: Approximate power consumption of 43557 Joules per run.3. EPYC 9554: Approximate power consumption of 37387 Joules per run.4. EPYC 9754: Approximate power consumption of 34726 Joules per run.5. EPYC 9684X: Approximate power consumption of 42393 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 46187 Joules per run.

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H13K26K39K52K65KSE +/- 15.14, N = 5SE +/- 537.84, N = 15SE +/- 363.39, N = 8SE +/- 391.92, N = 15SE +/- 469.70, N = 9SE +/- 268.67, N = 1517272.7354465.0445867.4444431.5859557.0835560.251. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H90180270360450116.77380.13332.67297.51437.49162.31
MinAvgMaxAmpere Altra Max 128c25.3147.9236.3EPYC 965425.9143.3319.2EPYC 955421.9137.9309.5EPYC 975418.7149.3321.4EPYC 9684X28.7136.1291.5Xeon Platinum 8490H54.8219.1377.0OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 2130 Joules per run.2. EPYC 9654: Approximate power consumption of 1118 Joules per run.3. EPYC 9554: Approximate power consumption of 1137 Joules per run.4. EPYC 9754: Approximate power consumption of 1424 Joules per run.5. EPYC 9684X: Approximate power consumption of 1044 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 1986 Joules per run.

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H3K6K9K12K15KSE +/- 0.36, N = 3SE +/- 32.54, N = 3SE +/- 9.54, N = 3SE +/- 9.06, N = 3SE +/- 240.55, N = 12SE +/- 4.50, N = 34295.7212332.4010327.5012660.8014731.508112.071. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Sysbench

This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPUAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H110K220K330K440K550KSE +/- 46.85, N = 3SE +/- 81.03, N = 3SE +/- 163.93, N = 3SE +/- 81.24, N = 3SE +/- 75.75, N = 3SE +/- 3.36, N = 3437954.92454951.33311423.41514483.58388317.80154451.201. (CC) gcc options: -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgEvents Per Second Per Watt, More Is BetterSysbench 1.0.20Test: CPUAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70014002100280035003487.461523.611359.831909.141645.67618.30
MinAvgMaxAmpere Altra Max 128c23.2125.6132.5EPYC 965427.5298.6317.2EPYC 955422.5229.0240.5EPYC 975480.7269.5281.0EPYC 9684X30.0236.0246.9Xeon Platinum 8490H59.4249.8259.7OpenBenchmarking.orgWatts, Fewer Is BetterSysbench 1.0.20CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 11805 Joules per run.2. EPYC 9654: Approximate power consumption of 27869 Joules per run.3. EPYC 9554: Approximate power consumption of 21451 Joules per run.4. EPYC 9754: Approximate power consumption of 25242 Joules per run.5. EPYC 9684X: Approximate power consumption of 22102 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 23481 Joules per run.

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 128Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H7001400210028003500SE +/- 0.84, N = 3SE +/- 1.99, N = 3SE +/- 3.26, N = 3SE +/- 8.00, N = 3SE +/- 10.12, N = 3SE +/- 2.52, N = 3940.82578.62400.42702.53102.41764.6-march=armv8.1-a-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.2-lquadmath -msse4.21. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden
OpenBenchmarking.orgGFLOPS/s Per Watt, More Is Betterlibxsmm 2-1.17-3645M N K: 128Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H481216207.63010.64410.72915.54412.4845.110
OpenBenchmarking.orgWatts, Fewer Is Betterlibxsmm 2-1.17-3645CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 23.27 / Avg: 123.31 / Max: 209.47Min: 27.33 / Avg: 242.26 / Max: 342.05Min: 22.5 / Avg: 223.74 / Max: 337.83Min: 18.29 / Avg: 173.86 / Max: 325.53Min: 30.86 / Avg: 248.52 / Max: 342.62Min: 59.22 / Avg: 345.34 / Max: 378.61. Ampere Altra Max 128c: Approximate power consumption of 60256 Joules per run.2. EPYC 9654: Approximate power consumption of 120242 Joules per run.3. EPYC 9554: Approximate power consumption of 104263 Joules per run.4. EPYC 9754: Approximate power consumption of 90697 Joules per run.5. EPYC 9684X: Approximate power consumption of 134863 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 102335 Joules per run.

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-256-GCMAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200000M400000M600000M800000M1000000MSE +/- 71755303.93, N = 3SE +/- 362414618.81, N = 3SE +/- 173027468.80, N = 3SE +/- 999773830.17, N = 3SE +/- 1296172747.45, N = 3SE +/- 1678798033.18, N = 33064922202107810247405136493858378171009634449730769778288320649241962540-lssl -lcrypto-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: AES-256-GCMAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H600M1200M1800M2400M3000M1491776544.172394262507.272014092746.632923318824.062351979989.351939199848.82
MinAvgMaxAmpere Altra Max 128c22.8205.5232.9EPYC 965428.0326.2348.9EPYC 955423.0322.4344.8EPYC 975481.2345.4361.1EPYC 9684X28.2327.3344.3Xeon Platinum 8490H73.4334.8376.8OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 37804 Joules per run.2. EPYC 9654: Approximate power consumption of 59696 Joules per run.3. EPYC 9554: Approximate power consumption of 59003 Joules per run.4. EPYC 9754: Approximate power consumption of 63203 Joules per run.5. EPYC 9684X: Approximate power consumption of 59894 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 61491 Joules per run.

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H612182430SE +/- 0.05179277, N = 3SE +/- 0.07670149, N = 5SE +/- 0.03040517, N = 5SE +/- 0.04030589, N = 5SE +/- 0.03937117, N = 5SE +/- 0.04713862, N = 423.885224008.452129369.745516219.028061687.3081441912.798342001. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
MinAvgMaxAmpere Altra Max 128c23.0129.8163.7EPYC 965426.2215.2324.4EPYC 955422.7184.0283.3EPYC 975479.0214.8306.9EPYC 9684X30.4206.1315.3Xeon Platinum 8490H65.3280.3374.7OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4112 Joules per run.2. EPYC 9654: Approximate power consumption of 2969 Joules per run.3. EPYC 9554: Approximate power consumption of 2686 Joules per run.4. EPYC 9754: Approximate power consumption of 3308 Joules per run.5. EPYC 9684X: Approximate power consumption of 2638 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 5115 Joules per run.

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30K60K90K120K150KSE +/- 10.27, N = 7SE +/- 1159.21, N = 15SE +/- 1712.13, N = 15SE +/- 667.76, N = 9SE +/- 1538.33, N = 15SE +/- 405.89, N = 942291.84118608.07125921.29127404.16137608.6888049.901. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500407.261167.531439.341333.531291.84549.13
MinAvgMaxAmpere Altra Max 128c23.3103.8194.0EPYC 965426.1101.6285.1EPYC 955421.087.5258.1EPYC 975418.595.5282.8EPYC 9684X27.6106.5285.1Xeon Platinum 8490H58.8160.3338.0OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 1038 Joules per run.2. EPYC 9654: Approximate power consumption of 677 Joules per run.3. EPYC 9554: Approximate power consumption of 566 Joules per run.4. EPYC 9754: Approximate power consumption of 743 Joules per run.5. EPYC 9684X: Approximate power consumption of 689 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 1105 Joules per run.

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500SE +/- 0.92, N = 3SE +/- 1.59, N = 3SE +/- 1.48, N = 3SE +/- 7.65, N = 3SE +/- 1.80, N = 3SE +/- 2.14, N = 31452.30665.04470.731037.89684.12447.51-mcpu=native - MIN: 1432.14-msse4.1 - MIN: 652.56-msse4.1 - MIN: 463.75-msse4.1 - MIN: 940.85-msse4.1 - MIN: 673.03-msse4.1 - MIN: 436.631. (CXX) g++ options: -O3 -march=native -fopenmp -fPIC -pie -ldl -lpthread
MinAvgMaxAmpere Altra Max 128c24.3178.6236.6EPYC 965426.9218.4300.7EPYC 955421.9196.8302.2EPYC 975478.2173.4291.3EPYC 9684X28.0222.9296.2Xeon Platinum 8490H75.5295.1375.2OpenBenchmarking.orgWatts, Fewer Is BetteroneDNN 3.3CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 15482 Joules per run.2. EPYC 9654: Approximate power consumption of 18416 Joules per run.3. EPYC 9554: Approximate power consumption of 16527 Joules per run.4. EPYC 9754: Approximate power consumption of 14968 Joules per run.5. EPYC 9684X: Approximate power consumption of 18800 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 24493 Joules per run.

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-128-GCMAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H300000M600000M900000M1200000M1500000MSE +/- 3971607.26, N = 3SE +/- 1300559723.82, N = 3SE +/- 4139145729.67, N = 3SE +/- 819456410.65, N = 3SE +/- 516635412.32, N = 3SE +/- 62680847.75, N = 33828502741009104169317737536518157631174244639723896750134797768644121760-lssl -lcrypto-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: AES-128-GCMAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H700M1400M2100M2800M3500M1810172827.552749579371.212306380943.303375536723.272715462059.952365142947.73
MinAvgMaxAmpere Altra Max 128c22.9211.5233.8EPYC 965428.0331.1355.1EPYC 955422.0326.8348.1EPYC 975481.8347.9363.2EPYC 9684X31.2330.2346.4Xeon Platinum 8490H68.0325.0377.3OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 38916 Joules per run.2. EPYC 9654: Approximate power consumption of 60593 Joules per run.3. EPYC 9554: Approximate power consumption of 59799 Joules per run.4. EPYC 9754: Approximate power consumption of 63660 Joules per run.5. EPYC 9684X: Approximate power consumption of 60434 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 59798 Joules per run.

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H15K30K45K60K75KSE +/- 0.46, N = 3SE +/- 405.08, N = 3SE +/- 219.20, N = 3SE +/- 655.18, N = 14SE +/- 485.47, N = 3SE +/- 146.74, N = 32416.5447003.2037265.9027269.2067965.5022365.201. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMpix/sec Per Watt, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H6012018024030017.51196.62191.48129.89275.7671.04
MinAvgMaxAmpere Altra Max 128c26.0138.0160.5EPYC 965427.1239.1335.0EPYC 955422.3194.6312.3EPYC 975458.7209.9325.3EPYC 9684X30.8246.5320.3Xeon Platinum 8490H63.7314.8359.0OpenBenchmarking.orgWatts, Fewer Is BetterASKAP 1.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 31657 Joules per run.2. EPYC 9654: Approximate power consumption of 10439 Joules per run.3. EPYC 9554: Approximate power consumption of 6812 Joules per run.4. EPYC 9754: Approximate power consumption of 17920 Joules per run.5. EPYC 9684X: Approximate power consumption of 9119 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 13433 Joules per run.

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Floating PointAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60K120K180K240K300KSE +/- 39.83, N = 3SE +/- 711.78, N = 3SE +/- 501.85, N = 3SE +/- 685.71, N = 3SE +/- 391.96, N = 3SE +/- 194.19, N = 3102559.01215564.78181377.97280779.02214129.85138052.19-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Vector Floating PointAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2004006008001000718.26761.61649.16916.63743.74601.07
MinAvgMaxAmpere Altra Max 128c23.0142.8191.1EPYC 965426.5283.0320.1EPYC 955423.0279.4317.4EPYC 975481.2306.3338.2EPYC 9684X31.3287.9327.4Xeon Platinum 8490H63.1229.7253.2OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 4902 Joules per run.2. EPYC 9654: Approximate power consumption of 9718 Joules per run.3. EPYC 9554: Approximate power consumption of 9500 Joules per run.4. EPYC 9754: Approximate power consumption of 10415 Joules per run.5. EPYC 9684X: Approximate power consumption of 9885 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 7886 Joules per run.

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA256Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H40000M80000M120000M160000M200000MSE +/- 29832593.30, N = 3SE +/- 45500182.34, N = 3SE +/- 58524233.88, N = 3SE +/- 83429444.82, N = 3SE +/- 33747498.40, N = 3SE +/- 52301160.39, N = 310116721810713035542227310753451759716532681505712459651775762987799233-lssl -lcrypto-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: SHA256Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H120M240M360M480M600M574282029.46403401627.27343065816.53496163836.00400002059.84192641019.65
MinAvgMaxAmpere Altra Max 128c21.4176.2210.5EPYC 965427.3323.1359.7EPYC 955422.7313.5355.1EPYC 975479.9333.2352.6EPYC 9684X30.3311.5335.3Xeon Platinum 8490H63.0327.0376.9OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 32414 Joules per run.2. EPYC 9654: Approximate power consumption of 59135 Joules per run.3. EPYC 9554: Approximate power consumption of 57362 Joules per run.4. EPYC 9754: Approximate power consumption of 60977 Joules per run.5. EPYC 9684X: Approximate power consumption of 57003 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 60053 Joules per run.

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression RatingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H170K340K510K680K850KSE +/- 159.75, N = 3SE +/- 609.70, N = 3SE +/- 599.23, N = 3SE +/- 957.95, N = 3SE +/- 3771.78, N = 3SE +/- 128.86, N = 35414646486305215697741226234892999021. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
OpenBenchmarking.orgMIPS Per Watt, More Is Better7-Zip Compression 22.01Test: Decompression RatingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H90018002700360045004112.432469.212097.033099.002440.92963.49
MinAvgMaxAmpere Altra Max 128c23.1131.7179.9EPYC 965426.9262.7348.0EPYC 955422.3248.7340.0EPYC 975480.2249.8342.7EPYC 9684X30.2255.4337.1Xeon Platinum 8490H72.6311.3376.8OpenBenchmarking.orgWatts, Fewer Is Better7-Zip Compression 22.01CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4959 Joules per run.2. EPYC 9654: Approximate power consumption of 10245 Joules per run.3. EPYC 9554: Approximate power consumption of 8042 Joules per run.4. EPYC 9754: Approximate power consumption of 11491 Joules per run.5. EPYC 9684X: Approximate power consumption of 10217 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 13799 Joules per run.

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H170K340K510K680K850KSE +/- 22.94, N = 3SE +/- 84.27, N = 3SE +/- 91.37, N = 3SE +/- 1066.79, N = 3SE +/- 77.65, N = 3SE +/- 127.07, N = 3399080.05611736.59507522.16773130.23585746.56315559.62-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Vector MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60012001800240030002938.302162.471804.272610.572126.531006.09
MinAvgMaxAmpere Altra Max 128c22.9135.8156.2EPYC 965428.0282.9322.3EPYC 955422.8281.3319.2EPYC 975480.9296.2330.4EPYC 9684X31.3275.4312.2Xeon Platinum 8490H63.3313.6351.0OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4663 Joules per run.2. EPYC 9654: Approximate power consumption of 9618 Joules per run.3. EPYC 9554: Approximate power consumption of 9658 Joules per run.4. EPYC 9754: Approximate power consumption of 10267 Joules per run.5. EPYC 9684X: Approximate power consumption of 9365 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10769 Joules per run.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2M4M6M8M10MSE +/- 234.11, N = 3SE +/- 527.77, N = 3SE +/- 3784.35, N = 3SE +/- 6288.41, N = 3SE +/- 7626.35, N = 3SE +/- 9446.68, N = 34692150.548271540.067240174.3611411104.748459429.384947898.66-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H8K16K24K32K40K25073.5430486.8426571.9236979.4329926.1419657.31
MinAvgMaxAmpere Altra Max 128c23.2187.1220.6EPYC 965427.7271.3306.9EPYC 955422.7272.5309.7EPYC 975480.8308.6343.3EPYC 9684X31.3282.7323.4Xeon Platinum 8490H70.8251.7338.3OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 6425 Joules per run.2. EPYC 9654: Approximate power consumption of 9315 Joules per run.3. EPYC 9554: Approximate power consumption of 9264 Joules per run.4. EPYC 9754: Approximate power consumption of 10595 Joules per run.5. EPYC 9684X: Approximate power consumption of 9705 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 8642 Joules per run.

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1020304050SE +/- 0.09, N = 4SE +/- 0.18, N = 7SE +/- 0.28, N = 6SE +/- 0.13, N = 7SE +/- 0.12, N = 7SE +/- 0.13, N = 518.1139.2830.1743.6840.3223.261. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H0.05020.10040.15060.20080.2510.1360.2200.1710.2230.2190.091
MinAvgMaxAmpere Altra Max 128c22.2133.0187.3EPYC 965422.4178.7303.4EPYC 955422.2176.8285.0EPYC 975478.6195.6328.6EPYC 9684X29.3183.7305.3Xeon Platinum 8490H60.8257.0377.3OpenBenchmarking.orgWatts, Fewer Is BetterACES DGEMM 1.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 2294 Joules per run.2. EPYC 9654: Approximate power consumption of 1787 Joules per run.3. EPYC 9554: Approximate power consumption of 2033 Joules per run.4. EPYC 9754: Approximate power consumption of 1845 Joules per run.5. EPYC 9684X: Approximate power consumption of 1811 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 3598 Joules per run.

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 8.0Length: 1e13Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1224364860SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 341.6824.1628.1421.7625.9752.381. (CXX) g++ options: -O3
MinAvgMaxAmpere Altra Max 128c22.7188.3222.8EPYC 965426.5308.6365.8EPYC 955422.6314.6361.0EPYC 975479.3277.7324.5EPYC 9684X30.5292.2343.3Xeon Platinum 8490H68.3329.4376.7OpenBenchmarking.orgWatts, Fewer Is BetterPrimesieve 8.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 8599 Joules per run.2. EPYC 9654: Approximate power consumption of 8640 Joules per run.3. EPYC 9554: Approximate power consumption of 10069 Joules per run.4. EPYC 9754: Approximate power consumption of 7128 Joules per run.5. EPYC 9684X: Approximate power consumption of 8766 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 18557 Joules per run.

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H11000M22000M33000M44000M55000MSE +/- 36142057.88, N = 3SE +/- 16069151.20, N = 3SE +/- 4732001.41, N = 3SE +/- 1530571.35, N = 3SE +/- 6401405.16, N = 3SE +/- 147259946.50, N = 3344349607974045932361033622028530536480782034010563958022445820447-lssl -lcrypto-m64-m64-m64-m64-m64 -lssl -lcrypto1. (CC) gcc options: -pthread -O3 -ldl
OpenBenchmarking.orgbyte/s Per Watt, More Is BetterOpenSSL 3.1Algorithm: SHA512Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H40M80M120M160M200M192898458.48133650046.77112644110.72160226634.86131019550.9165707856.51
MinAvgMaxAmpere Altra Max 128c23.0178.5195.4EPYC 965428.0302.7332.5EPYC 955423.0298.5327.7EPYC 975481.3334.8349.1EPYC 9684X31.4306.1329.5Xeon Platinum 8490H58.7341.6377.3OpenBenchmarking.orgWatts, Fewer Is BetterOpenSSL 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 32846 Joules per run.2. EPYC 9654: Approximate power consumption of 55399 Joules per run.3. EPYC 9554: Approximate power consumption of 54622 Joules per run.4. EPYC 9754: Approximate power consumption of 61385 Joules per run.5. EPYC 9684X: Approximate power consumption of 56119 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 62854 Joules per run.

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H20406080100SE +/- 0.79, N = 3SE +/- 0.46, N = 4SE +/- 0.39, N = 4SE +/- 0.44, N = 5SE +/- 0.45, N = 4SE +/- 0.49, N = 378.5536.6033.5241.5536.5935.60
MinAvgMaxAmpere Altra Max 128c22.2124.2181.6EPYC 965426.9174.6309.7EPYC 955421.9157.4318.6EPYC 975477.7152.5294.4EPYC 9684X29.5177.0299.5Xeon Platinum 8490H67.9230.6376.5OpenBenchmarking.orgWatts, Fewer Is BetterTimed Linux Kernel Compilation 6.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 11343 Joules per run.2. EPYC 9654: Approximate power consumption of 8033 Joules per run.3. EPYC 9554: Approximate power consumption of 6767 Joules per run.4. EPYC 9754: Approximate power consumption of 7871 Joules per run.5. EPYC 9684X: Approximate power consumption of 8186 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10529 Joules per run.

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H50100150200250SE +/- 0.35, N = 3SE +/- 0.16, N = 3SE +/- 0.03, N = 3SE +/- 0.26, N = 3SE +/- 0.06, N = 3SE +/- 0.43, N = 3227.78101.56103.50118.2598.56128.47
MinAvgMaxAmpere Altra Max 128c24.494.2182.8EPYC 965426.9181.4340.6EPYC 955422.0178.6328.9EPYC 975477.1156.8333.8EPYC 9684X27.3188.9334.5Xeon Platinum 8490H66.8258.4373.6OpenBenchmarking.orgWatts, Fewer Is BetterTimed Godot Game Engine Compilation 4.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 22010 Joules per run.2. EPYC 9654: Approximate power consumption of 19466 Joules per run.3. EPYC 9554: Approximate power consumption of 19464 Joules per run.4. EPYC 9754: Approximate power consumption of 19446 Joules per run.5. EPYC 9684X: Approximate power consumption of 19643 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 34533 Joules per run.

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Random ReadAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H130M260M390M520M650MSE +/- 3039704.34, N = 15SE +/- 1006626.36, N = 3SE +/- 106225.72, N = 3SE +/- 817605.37, N = 3SE +/- 539077.40, N = 3SE +/- 6039899.37, N = 154354372764293102633769834156112591614868627852683744861. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s Per Watt, More Is BetterRocksDB 8.0Test: Random ReadAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H500K1000K1500K2000K2500K2282816.531599837.631450487.412068476.161587869.93808152.35
MinAvgMaxAmpere Altra Max 128c23.1190.7208.7EPYC 965427.2268.3293.5EPYC 955422.2259.9283.8EPYC 975479.0295.5314.1EPYC 9684X30.9306.6327.5Xeon Platinum 8490H59.2332.1378.9OpenBenchmarking.orgWatts, Fewer Is BetterRocksDB 8.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 12220 Joules per run.2. EPYC 9654: Approximate power consumption of 17174 Joules per run.3. EPYC 9554: Approximate power consumption of 16634 Joules per run.4. EPYC 9754: Approximate power consumption of 18913 Joules per run.5. EPYC 9684X: Approximate power consumption of 19521 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 21276 Joules per run.

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H120240360480600SE +/- 1.09, N = 3SE +/- 4.91, N = 3SE +/- 0.48, N = 3SE +/- 1.09, N = 3SE +/- 8.88, N = 3SE +/- 3.94, N = 3253.26554.30571.53445.46558.78487.01MIN: 27.19 / MAX: 2608.7MIN: 57.47 / MAX: 5454.55MIN: 51.5 / MAX: 5454.55MIN: 63.69 / MAX: 3750MIN: 56.02 / MAX: 5454.55MIN: 40.6 / MAX: 5454.55
OpenBenchmarking.orgQueries Per Minute, Geo Mean Per Watt, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1.08952.1793.26854.3585.44753.8804.0574.8423.9853.8062.259
MinAvgMaxAmpere Altra Max 128c23.665.3177.9EPYC 965426.5136.6303.6EPYC 955420.7118.0308.6EPYC 975452.3111.8267.4EPYC 9684X29.7146.8290.3Xeon Platinum 8490H72.4215.6362.5OpenBenchmarking.orgWatts, Fewer Is BetterClickHouse 22.12.3.5CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 19536 Joules per run.2. EPYC 9654: Approximate power consumption of 25593 Joules per run.3. EPYC 9554: Approximate power consumption of 21640 Joules per run.4. EPYC 9754: Approximate power consumption of 24816 Joules per run.5. EPYC 9684X: Approximate power consumption of 28192 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 42477 Joules per run.

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To CompileAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60120180240300SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.32, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.27, N = 3268.70120.16121.93130.46119.20151.57
MinAvgMaxAmpere Altra Max 128c22.0119.2178.8EPYC 965427.1221.5344.1EPYC 955422.4215.2336.3EPYC 975478.6196.8336.6EPYC 9684X30.8224.1336.0Xeon Platinum 8490H67.8290.0376.5OpenBenchmarking.orgWatts, Fewer Is BetterTimed Node.js Compilation 19.8.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 33142 Joules per run.2. EPYC 9654: Approximate power consumption of 28350 Joules per run.3. EPYC 9554: Approximate power consumption of 27910 Joules per run.4. EPYC 9754: Approximate power consumption of 27358 Joules per run.5. EPYC 9684X: Approximate power consumption of 28536 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 46211 Joules per run.

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H150K300K450K600K750KSE +/- 389.46, N = 3SE +/- 508.29, N = 3SE +/- 6782.96, N = 12SE +/- 236.23, N = 3SE +/- 944.14, N = 3SE +/- 3004.11, N = 6682356.13421053.02346518.02552067.04417406.39304345.30-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Matrix MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70014002100280035003471.891484.811235.581768.351435.57962.55
MinAvgMaxAmpere Altra Max 128c23.0196.5241.5EPYC 965427.7283.6323.6EPYC 955422.9280.4318.7EPYC 975480.5312.2348.2EPYC 9684X29.4290.8329.0Xeon Platinum 8490H58.1316.2376.7OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 6748 Joules per run.2. EPYC 9654: Approximate power consumption of 9736 Joules per run.3. EPYC 9554: Approximate power consumption of 9559 Joules per run.4. EPYC 9754: Approximate power consumption of 10719 Joules per run.5. EPYC 9684X: Approximate power consumption of 9983 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10803 Joules per run.

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H500M1000M1500M2000M2500MSE +/- 31942.66, N = 3SE +/- 5502047.33, N = 3SE +/- 3180941.37, N = 3SE +/- 2811300.43, N = 3SE +/- 1953221.73, N = 3SE +/- 269139.99, N = 31060656000229673066723211560002291049667237485333316118260001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
OpenBenchmarking.orgFigure Of Merit Per Watt, More Is BetterAlgebraic Multi-Grid Benchmark 1.2Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H3M6M9M12M15M7481610.119775985.2012260061.4911292392.9310140759.125406835.94
MinAvgMaxAmpere Altra Max 128c23.1141.8169.0EPYC 965427.2234.9304.8EPYC 955422.2189.3270.8EPYC 975418.9202.9293.2EPYC 9684X30.8234.2299.7Xeon Platinum 8490H68.7298.1368.7OpenBenchmarking.orgWatts, Fewer Is BetterAlgebraic Multi-Grid Benchmark 1.2CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 10727 Joules per run.2. EPYC 9654: Approximate power consumption of 7988 Joules per run.3. EPYC 9554: Approximate power consumption of 4481 Joules per run.4. EPYC 9754: Approximate power consumption of 9197 Joules per run.5. EPYC 9684X: Approximate power consumption of 7884 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 8248 Joules per run.

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60M120M180M240M300MSE +/- 691803.98, N = 3SE +/- 612439.42, N = 3SE +/- 662770.52, N = 3SE +/- 2513657.44, N = 3SE +/- 906405.74, N = 3SE +/- 819760.48, N = 3212432479250209020210167054290260505233926631130818594
OpenBenchmarking.orgNodes/second Per Watt, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H300K600K900K1200K1500K1179451.42776992.64670647.64920803.01756672.13383452.52
MinAvgMaxAmpere Altra Max 128c24.0180.1193.8EPYC 965426.7322.0347.3EPYC 955422.3313.4340.0EPYC 975481.5315.2339.8EPYC 9684X29.8309.2329.2Xeon Platinum 8490H66.0341.2375.8OpenBenchmarking.orgWatts, Fewer Is BetterasmFish 2018-07-23CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 18131 Joules per run.2. EPYC 9654: Approximate power consumption of 40145 Joules per run.3. EPYC 9554: Approximate power consumption of 30711 Joules per run.4. EPYC 9754: Approximate power consumption of 41189 Joules per run.5. EPYC 9684X: Approximate power consumption of 38953 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 51515 Joules per run.

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H11K22K33K44K55KSE +/- 5.88, N = 4SE +/- 45.75, N = 5SE +/- 51.65, N = 5SE +/- 105.96, N = 5SE +/- 507.76, N = 5SE +/- 14.30, N = 424122.050320.250351.450393.752937.635452.51. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi
OpenBenchmarking.orgCG Mflops Per Watt, More Is BetterminiFE 2.2Problem Size: SmallAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H90180270360450248.68340.61396.88416.05351.89154.24
MinAvgMaxAmpere Altra Max 128c23.397.0166.1EPYC 965425.5147.7292.0EPYC 955421.9126.9255.2EPYC 975417.7121.1251.2EPYC 9684X28.7150.4291.8Xeon Platinum 8490H60.0229.9376.4OpenBenchmarking.orgWatts, Fewer Is BetterminiFE 2.2CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 1964 Joules per run.2. EPYC 9654: Approximate power consumption of 2068 Joules per run.3. EPYC 9554: Approximate power consumption of 1751 Joules per run.4. EPYC 9754: Approximate power consumption of 1793 Joules per run.5. EPYC 9684X: Approximate power consumption of 2076 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 3620 Joules per run.

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H40K80K120K160K200KSE +/- 271.90, N = 3SE +/- 69.15, N = 3SE +/- 33.83, N = 3SE +/- 152.02, N = 3SE +/- 0.67, N = 3SE +/- 30.49, N = 310875316685013803520482816611993877-m64-m64-m64-m64-m641. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: bcryptAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H150300450600750715.08614.12526.01711.37606.08301.12
MinAvgMaxAmpere Altra Max 128c23.6152.1177.4EPYC 965426.7271.7308.3EPYC 955422.2262.4299.1EPYC 975478.7287.9323.4EPYC 9684X29.4274.1313.0Xeon Platinum 8490H59.9311.8348.7OpenBenchmarking.orgWatts, Fewer Is BetterJohn The Ripper 2023.03.14CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 5222 Joules per run.2. EPYC 9654: Approximate power consumption of 9237 Joules per run.3. EPYC 9554: Approximate power consumption of 9010 Joules per run.4. EPYC 9754: Approximate power consumption of 9886 Joules per run.5. EPYC 9684X: Approximate power consumption of 9410 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10704 Joules per run.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H40K80K120K160K200KSE +/- 275.70, N = 3SE +/- 18.68, N = 3SE +/- 11.89, N = 3SE +/- 493.13, N = 3SE +/- 18.85, N = 3SE +/- 10.67, N = 310882716688713803220467716608093913-m64-m64-m64-m64-m641. (CC) gcc options: -lssl -lcrypto -fopenmp -lgmp -lm -lrt -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Watt, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H160320480640800720.26615.51520.18699.30601.89300.62

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H120240360480600SE +/- 1.12, N = 3SE +/- 1.65, N = 3SE +/- 2.76, N = 3SE +/- 0.34, N = 3SE +/- 5.05, N = 3SE +/- 4.34, N = 3253.53551.97546.67431.65546.14485.91MIN: 29.51 / MAX: 3157.89MIN: 57.47 / MAX: 4285.71MIN: 51.24 / MAX: 3529.41MIN: 59.06 / MAX: 3157.89MIN: 55.87 / MAX: 5000MIN: 40.11 / MAX: 5000

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 4.1.3Test: WritesAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60K120K180K240K300KSE +/- 2776.53, N = 3SE +/- 162.88, N = 3SE +/- 554.18, N = 3SE +/- 1223.06, N = 3SE +/- 1195.73, N = 3SE +/- 883.78, N = 3227548285324301963236320286460139654
OpenBenchmarking.orgOp/s Per Watt, More Is BetterApache Cassandra 4.1.3Test: WritesAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H4008001200160020002068.001349.721700.281432.671282.09557.85
MinAvgMaxAmpere Altra Max 128c23.2110.0133.2EPYC 965426.8211.4268.5EPYC 955422.7177.6215.1EPYC 975479.3165.0203.6EPYC 9684X31.5223.4260.5Xeon Platinum 8490H77.9250.3325.8OpenBenchmarking.orgWatts, Fewer Is BetterApache Cassandra 4.1.3CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 16101 Joules per run.2. EPYC 9654: Approximate power consumption of 30793 Joules per run.3. EPYC 9554: Approximate power consumption of 25811 Joules per run.4. EPYC 9754: Approximate power consumption of 24083 Joules per run.5. EPYC 9684X: Approximate power consumption of 32547 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 36133 Joules per run.

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H120240360480600SE +/- 2.89, N = 3SE +/- 2.48, N = 3SE +/- 3.20, N = 3SE +/- 2.30, N = 3SE +/- 0.68, N = 3SE +/- 5.21, N = 3250.30531.52535.55425.51534.25461.90MIN: 31.02 / MAX: 3000MIN: 56.87 / MAX: 6666.67MIN: 50.63 / MAX: 6000MIN: 48.54 / MAX: 3333.33MIN: 54.1 / MAX: 4285.71MIN: 40.49 / MAX: 5454.55

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H800K1600K2400K3200K4000KSE +/- 21790.17, N = 12SE +/- 42157.02, N = 3SE +/- 10640.30, N = 3SE +/- 22337.40, N = 3SE +/- 36381.27, N = 3SE +/- 10652.33, N = 32123217368857935156233461157352594517322131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average LatencyAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H0.12980.25960.38940.51920.649SE +/- 0.005, N = 12SE +/- 0.003, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 3SE +/- 0.003, N = 3SE +/- 0.004, N = 30.4720.2710.2840.2890.2830.5771. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 16CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 21.77 / Avg: 156.33 / Max: 196.7Min: 25.08 / Avg: 273.19 / Max: 320.26Min: 21.94 / Avg: 260.1 / Max: 304.42Min: 77.1 / Avg: 252.99 / Max: 315.23Min: 29.11 / Avg: 274.69 / Max: 317.84Min: 61.4 / Avg: 308.67 / Max: 366.131. Ampere Altra Max 128c: Approximate power consumption of 22902 Joules per run.2. EPYC 9654: Approximate power consumption of 38793 Joules per run.3. EPYC 9554: Approximate power consumption of 36934 Joules per run.4. EPYC 9754: Approximate power consumption of 36263 Joules per run.5. EPYC 9684X: Approximate power consumption of 38914 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 43522 Joules per run.

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1000K2000K3000K4000K5000KSE +/- 1388.32, N = 3SE +/- 9714.51, N = 3SE +/- 19980.87, N = 3SE +/- 25359.52, N = 3SE +/- 5924.03, N = 3SE +/- 1523.22, N = 32829772.923884296.773056532.144543138.123703841.722233945.611. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec Per Watt, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H6K12K18K24K30K28735.3513250.6410800.4016505.7713223.677546.97
MinAvgMaxAmpere Altra Max 128c22.998.5158.5EPYC 965427.6293.1351.9EPYC 955422.4283.0328.6EPYC 975480.6275.2351.7EPYC 9684X31.6280.1326.7Xeon Platinum 8490H59.9296.0342.1OpenBenchmarking.orgWatts, Fewer Is BetterCoremark 1.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 2232 Joules per run.2. EPYC 9654: Approximate power consumption of 13680 Joules per run.3. EPYC 9554: Approximate power consumption of 16225 Joules per run.4. EPYC 9754: Approximate power consumption of 11469 Joules per run.5. EPYC 9684X: Approximate power consumption of 13725 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 14998 Joules per run.

Timed LLVM Compilation

This test times how long it takes to compile/build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60120180240300SE +/- 0.95, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.28, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3266.54133.07131.20146.39131.99169.53
MinAvgMaxAmpere Altra Max 128c22.9118.4179.2EPYC 965427.1238.4333.5EPYC 955422.7235.5327.1EPYC 975478.8217.7330.7EPYC 9684X30.6239.8329.2Xeon Platinum 8490H70.7300.4377.1OpenBenchmarking.orgWatts, Fewer Is BetterTimed LLVM Compilation 16.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 33897 Joules per run.2. EPYC 9654: Approximate power consumption of 35758 Joules per run.3. EPYC 9554: Approximate power consumption of 34930 Joules per run.4. EPYC 9754: Approximate power consumption of 36133 Joules per run.5. EPYC 9684X: Approximate power consumption of 35736 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 55477 Joules per run.

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H15K30K45K60K75KSE +/- 16.52, N = 3SE +/- 142.78, N = 3SE +/- 42.25, N = 3SE +/- 2550.91, N = 12SE +/- 80.10, N = 4SE +/- 7.51, N = 34322.159047.747926.429356.169205.627608.5-maes-maes-maes-maes-maes1. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s Per Watt, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H5010015020025036.63208.70166.56124.40245.1186.36
MinAvgMaxAmpere Altra Max 128c23.2118.0132.8EPYC 965426.6282.9358.5EPYC 955422.4287.7349.8EPYC 975478.8236.0328.4EPYC 9684X29.9281.5361.3Xeon Platinum 8490H60.3319.7377.1OpenBenchmarking.orgWatts, Fewer Is BetterXmrig 6.18.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 28086 Joules per run.2. EPYC 9654: Approximate power consumption of 6224 Joules per run.3. EPYC 9554: Approximate power consumption of 7481 Joules per run.4. EPYC 9754: Approximate power consumption of 9695 Joules per run.5. EPYC 9684X: Approximate power consumption of 5488 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 13213 Joules per run.

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H12K24K36K48K60KSE +/- 4.44, N = 3SE +/- 475.58, N = 3SE +/- 153.57, N = 3SE +/- 574.27, N = 15SE +/- 0.00, N = 3SE +/- 131.23, N = 35285.3641081.6031177.2026593.4058310.1019616.001. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H40K80K120K160K200KSE +/- 31.78, N = 3SE +/- 250.78, N = 4SE +/- 359.87, N = 4SE +/- 303.64, N = 4SE +/- 1294.93, N = 6SE +/- 69.75, N = 320218.43125373.87119472.30131257.15208538.3272703.351. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.5
OpenBenchmarking.orgTotal Mop/s Per Watt, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2004006008001000104.42550.81586.74644.871032.65244.61
MinAvgMaxAmpere Altra Max 128c23.2193.6212.3EPYC 965426.4227.6323.4EPYC 955422.2203.6309.7EPYC 975418.3203.5315.1EPYC 9684X29.0201.9317.5Xeon Platinum 8490H61.1297.2368.6OpenBenchmarking.orgWatts, Fewer Is BetterNAS Parallel Benchmarks 3.4CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 15038 Joules per run.2. EPYC 9654: Approximate power consumption of 3813 Joules per run.3. EPYC 9554: Approximate power consumption of 3512 Joules per run.4. EPYC 9754: Approximate power consumption of 3460 Joules per run.5. EPYC 9684X: Approximate power consumption of 2491 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 7530 Joules per run.

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Wide Vector MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H900K1800K2700K3600K4500KSE +/- 465.61, N = 3SE +/- 41.80, N = 3SE +/- 1186.05, N = 3SE +/- 495.80, N = 3SE +/- 30.71, N = 3SE +/- 3353.99, N = 32355041.962597630.052250265.273437154.962597371.594423715.19-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Wide Vector MathAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H3K6K9K12K15K13527.039120.777880.5211451.148987.0214009.41
MinAvgMaxAmpere Altra Max 128c23.3174.1205.6EPYC 965427.9284.8323.2EPYC 955422.8285.5323.3EPYC 975480.5300.2335.7EPYC 9684X31.2289.0328.2Xeon Platinum 8490H63.1315.8377.1OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 5919 Joules per run.2. EPYC 9654: Approximate power consumption of 9778 Joules per run.3. EPYC 9554: Approximate power consumption of 9709 Joules per run.4. EPYC 9754: Approximate power consumption of 10305 Joules per run.5. EPYC 9684X: Approximate power consumption of 9826 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10841 Joules per run.

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2M4M6M8M10MSE +/- 59438.15, N = 4SE +/- 8234.52, N = 3SE +/- 10341.18, N = 3SE +/- 18045.69, N = 3SE +/- 52827.93, N = 3SE +/- 48574.04, N = 154860402.438412361.968108606.407669596.617930429.134380269.261. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec Per Watt, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H8K16K24K32K40K35807.8131458.4834842.9233199.6530639.6915039.41
MinAvgMaxAmpere Altra Max 128c23.1135.7162.9EPYC 965427.5267.4310.6EPYC 955422.4232.7271.8EPYC 975478.5231.0259.6EPYC 9684X29.5258.8301.6Xeon Platinum 8490H56.7291.3336.5OpenBenchmarking.orgWatts, Fewer Is BetterMemcached 1.6.19CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 9637 Joules per run.2. EPYC 9654: Approximate power consumption of 18630 Joules per run.3. EPYC 9554: Approximate power consumption of 16290 Joules per run.4. EPYC 9754: Approximate power consumption of 16094 Joules per run.5. EPYC 9684X: Approximate power consumption of 18118 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 20368 Joules per run.

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SharpenAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500SE +/- 1.20, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 31274809669924775691-lxml2-lxml2-lxml2-lxml2-lxml21. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lzstd -lm -lpthread
OpenBenchmarking.orgIterations Per Minute Per Watt, More Is BetterGraphicsMagick 1.3.38Operation: SharpenAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2468108.0163.1272.6913.7523.0422.081
MinAvgMaxAmpere Altra Max 128c22.9158.9187.9EPYC 965427.5258.7278.7EPYC 955422.6248.6266.2EPYC 975479.9246.3262.1EPYC 9684X30.7254.7271.9Xeon Platinum 8490H61.6332.0369.6OpenBenchmarking.orgWatts, Fewer Is BetterGraphicsMagick 1.3.38CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 10225 Joules per run.2. EPYC 9654: Approximate power consumption of 16470 Joules per run.3. EPYC 9554: Approximate power consumption of 15911 Joules per run.4. EPYC 9754: Approximate power consumption of 15679 Joules per run.5. EPYC 9684X: Approximate power consumption of 16218 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 21137 Joules per run.

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60012001800240030002699.01568.41452.11700.81641.62056.6-m64-m64-m64-m64-m641. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2
OpenBenchmarking.orgWatts, Fewer Is BetterNWChem 7.0.2CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 23.24 / Avg: 198.82 / Max: 228.12Min: 27.56 / Avg: 342.56 / Max: 352.74Min: 22.2 / Avg: 330.83 / Max: 345.41Min: 17.77 / Avg: 343.89 / Max: 353.98Min: 30.42 / Avg: 323.18 / Max: 332.59Min: 70.91 / Avg: 347.65 / Max: 373.16

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Memory CopyingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H7K14K21K28K35KSE +/- 0.31, N = 3SE +/- 22.89, N = 3SE +/- 19.64, N = 3SE +/- 23.24, N = 3SE +/- 12.67, N = 3SE +/- 67.98, N = 327163.4528310.8222827.2034497.5126495.6318809.55-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Memory CopyingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H4080120160200166.5895.6078.72123.6797.0359.82
MinAvgMaxAmpere Altra Max 128c23.5163.1191.1EPYC 965426.6296.1337.4EPYC 955422.4290.0331.9EPYC 975479.2279.0314.4EPYC 9684X30.7273.1309.4Xeon Platinum 8490H58.2314.5376.1OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 5599 Joules per run.2. EPYC 9654: Approximate power consumption of 10168 Joules per run.3. EPYC 9554: Approximate power consumption of 9956 Joules per run.4. EPYC 9754: Approximate power consumption of 9577 Joules per run.5. EPYC 9684X: Approximate power consumption of 9285 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10797 Joules per run.

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Read While WritingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H2M4M6M8M10MSE +/- 89314.22, N = 15SE +/- 76682.69, N = 3SE +/- 76607.71, N = 5SE +/- 28493.97, N = 3SE +/- 90164.28, N = 3SE +/- 46694.84, N = 387919119704119741265011046045951845260911181. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s Per Watt, More Is BetterRocksDB 8.0Test: Read While WritingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H12K24K36K48K60K54308.8133265.3929788.4743697.4633658.0918426.07
MinAvgMaxAmpere Altra Max 128c23.4161.9195.2EPYC 965426.4291.7335.5EPYC 955422.5248.8327.1EPYC 975480.3252.8333.2EPYC 9684X31.3282.8340.7Xeon Platinum 8490H58.4330.6375.1OpenBenchmarking.orgWatts, Fewer Is BetterRocksDB 8.0CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 10372 Joules per run.2. EPYC 9654: Approximate power consumption of 18573 Joules per run.3. EPYC 9554: Approximate power consumption of 15926 Joules per run.4. EPYC 9754: Approximate power consumption of 16178 Joules per run.5. EPYC 9684X: Approximate power consumption of 18099 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 21267 Joules per run.

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H130K260K390K520K650KSE +/- 1588.70, N = 3SE +/- 311.94, N = 3SE +/- 224.17, N = 3SE +/- 439.85, N = 3SE +/- 366.95, N = 3SE +/- 629.56, N = 33283785649504897365865715544113896901. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H20406080100SE +/- 0.07, N = 3SE +/- 0.32, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.27, N = 3SE +/- 0.43, N = 356.9061.5371.2748.9564.3586.761. (CC) gcc options: -O2 -pthread
MinAvgMaxAmpere Altra Max 128c23.2221.4246.2EPYC 965427.7315.0343.8EPYC 955423.0315.2344.4EPYC 975480.1308.7344.3EPYC 9684X31.4303.2329.7Xeon Platinum 8490H68.9335.6377.4OpenBenchmarking.orgWatts, Fewer Is BetterHelsing 1.0-betaCPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 13433 Joules per run.2. EPYC 9654: Approximate power consumption of 20475 Joules per run.3. EPYC 9554: Approximate power consumption of 23641 Joules per run.4. EPYC 9754: Approximate power consumption of 16260 Joules per run.5. EPYC 9684X: Approximate power consumption of 20615 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 30542 Joules per run.

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Fused Multiply-AddAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H50M100M150M200M250MSE +/- 11189.02, N = 3SE +/- 301328.09, N = 3SE +/- 1366328.72, N = 3SE +/- 2076871.25, N = 3SE +/- 140980.34, N = 3SE +/- 222901.64, N = 3151397996.39202941117.45167642337.90256387572.72202387450.78145542113.66-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Fused Multiply-AddAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200K400K600K800K1000K750899.23694638.83582861.50864910.19681791.81462151.13
MinAvgMaxAmpere Altra Max 128c23.2201.6236.3EPYC 965427.8292.2333.6EPYC 955422.8287.6327.9EPYC 975480.9296.4331.8EPYC 9684X31.2296.8337.5Xeon Platinum 8490H64.9314.9355.4OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 6855 Joules per run.2. EPYC 9654: Approximate power consumption of 9933 Joules per run.3. EPYC 9554: Approximate power consumption of 9779 Joules per run.4. EPYC 9754: Approximate power consumption of 10276 Joules per run.5. EPYC 9684X: Approximate power consumption of 10192 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10812 Joules per run.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Floating PointAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H8K16K24K32K40KSE +/- 38.91, N = 3SE +/- 41.92, N = 3SE +/- 2.37, N = 3SE +/- 56.54, N = 3SE +/- 26.43, N = 3SE +/- 13.87, N = 322187.7230824.9724168.3639031.6429220.7822454.70-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: Floating PointAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H4080120160200176.29105.0387.19124.72105.1572.07
MinAvgMaxAmpere Altra Max 128c22.6125.9147.1EPYC 965426.3293.5332.7EPYC 955422.7277.2314.1EPYC 975479.3312.9348.7EPYC 9684X31.3277.9316.9Xeon Platinum 8490H65.8311.6347.7OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4279 Joules per run.2. EPYC 9654: Approximate power consumption of 9978 Joules per run.3. EPYC 9554: Approximate power consumption of 9425 Joules per run.4. EPYC 9754: Approximate power consumption of 10744 Joules per run.5. EPYC 9684X: Approximate power consumption of 9448 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10698 Joules per run.

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H80M160M240M320M400M2240320003911670003680160003770330003837540003331290001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H130260390520650SE +/- 0.07, N = 3SE +/- 8.73, N = 9SE +/- 7.98, N = 9SE +/- 14.11, N = 3SE +/- 5.21, N = 9SE +/- 4.12, N = 4606.52430.08515.79493.50440.55353.561. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
OpenBenchmarking.orgWatts, Fewer Is BetterXcompact3d Incompact3d 2021-03-11CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60120180240300Min: 23.24 / Avg: 155.04 / Max: 166.82Min: 26.82 / Avg: 300.11 / Max: 330.92Min: 22.74 / Avg: 236.48 / Max: 261.22Min: 18.66 / Avg: 263.92 / Max: 306.42Min: 30.14 / Avg: 300.86 / Max: 318.85Min: 74.07 / Avg: 345.25 / Max: 357.571. Ampere Altra Max 128c: Approximate power consumption of 96330 Joules per run.2. EPYC 9654: Approximate power consumption of 131250 Joules per run.3. EPYC 9554: Approximate power consumption of 123521 Joules per run.4. EPYC 9754: Approximate power consumption of 147794 Joules per run.5. EPYC 9684X: Approximate power consumption of 134718 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 124895 Joules per run.

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1020304050SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.53, N = 331.7930.2434.0125.1530.3242.711. (CXX) g++ options: -O2 -lOpenCL
MinAvgMaxAmpere Altra Max 128c23.2133.1212.1EPYC 965426.4247.8323.8EPYC 955421.9252.5321.0EPYC 975418.9226.2330.2EPYC 9684X30.6251.6321.7Xeon Platinum 8490H67.6290.0377.5OpenBenchmarking.orgWatts, Fewer Is BetterRodinia 3.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 4790 Joules per run.2. EPYC 9654: Approximate power consumption of 8427 Joules per run.3. EPYC 9554: Approximate power consumption of 9510 Joules per run.4. EPYC 9754: Approximate power consumption of 6635 Joules per run.5. EPYC 9684X: Approximate power consumption of 8554 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 13437 Joules per run.

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: TPC-H ParquetAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H50100150200250SE +/- 4.97, N = 9SE +/- 0.47, N = 3SE +/- 0.53, N = 3SE +/- 0.64, N = 3SE +/- 0.26, N = 3SE +/- 0.30, N = 3237.16146.28143.31177.13147.49148.101. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl
OpenBenchmarking.orgWatts, Fewer Is BetterDuckDB 0.9.1CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H50100150200250Min: 23 / Avg: 41.68 / Max: 67.92Min: 26.61 / Avg: 104.46 / Max: 171.82Min: 21.96 / Avg: 86.81 / Max: 156.77Min: 76.23 / Avg: 86.82 / Max: 128.75Min: 28.74 / Avg: 115.53 / Max: 185.42Min: 60.54 / Avg: 169.88 / Max: 268.131. Ampere Altra Max 128c: Approximate power consumption of 11352 Joules per run.2. EPYC 9654: Approximate power consumption of 20823 Joules per run.3. EPYC 9554: Approximate power consumption of 17015 Joules per run.4. EPYC 9754: Approximate power consumption of 20895 Joules per run.5. EPYC 9684X: Approximate power consumption of 23299 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 34428 Joules per run.

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H5K10K15K20K25KSE +/- 24.21, N = 3SE +/- 84.04, N = 3SE +/- 113.24, N = 3SE +/- 59.94, N = 3SE +/- 90.30, N = 3SE +/- 49.50, N = 516218.3323701.4424037.4422356.7524712.8823997.721. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
OpenBenchmarking.orgz/s Per Watt, More Is BetterLULESH 2.0.3Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H306090120150144.69136.97148.93124.36139.86106.69
MinAvgMaxAmpere Altra Max 128c23.4112.1165.3EPYC 965427.5173.0275.8EPYC 955422.1161.4260.1EPYC 975479.9179.8295.1EPYC 9684X29.9176.7279.0Xeon Platinum 8490H66.0224.9333.2OpenBenchmarking.orgWatts, Fewer Is BetterLULESH 2.0.3CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 6613 Joules per run.2. EPYC 9654: Approximate power consumption of 4268 Joules per run.3. EPYC 9554: Approximate power consumption of 3874 Joules per run.4. EPYC 9754: Approximate power consumption of 8509 Joules per run.5. EPYC 9684X: Approximate power consumption of 4300 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 2834 Joules per run.

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H714212835SE +/- 0.00, N = 3SE +/- 0.35, N = 3SE +/- 0.04, N = 3SE +/- 0.27, N = 3SE +/- 0.28, N = 9SE +/- 0.11, N = 321.2432.1322.4525.8923.7831.241. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Watt, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H0.02970.05940.08910.11880.14850.1320.1170.1010.1090.0870.091
OpenBenchmarking.orgWatts, Fewer Is BetterHigh Performance Conjugate Gradient 3.1CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60120180240300Min: 21.54 / Avg: 161.52 / Max: 176.13Min: 27.26 / Avg: 273.62 / Max: 311.89Min: 22.65 / Avg: 222.24 / Max: 257.64Min: 18.61 / Avg: 236.52 / Max: 294.94Min: 29.6 / Avg: 272.17 / Max: 306.17Min: 64.77 / Avg: 344.51 / Max: 354.151. Ampere Altra Max 128c: Approximate power consumption of 190213 Joules per run.2. EPYC 9654: Approximate power consumption of 180865 Joules per run.3. EPYC 9554: Approximate power consumption of 119416 Joules per run.4. EPYC 9754: Approximate power consumption of 229347 Joules per run.5. EPYC 9684X: Approximate power consumption of 212926 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 130684 Joules per run.

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H110M220M330M440M550M3332390004993470004674530005021510005012220004544830001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.orgsssp max_TEPS Per Watt, More Is BetterGraph500 3.0Scale: 26Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H400K800K1200K1600K2000K1918278.631644954.691818554.401711121.291663398.871312538.58
MinAvgMaxAmpere Altra Max 128c22.0173.7182.3EPYC 965426.9303.6343.1EPYC 955422.0257.0304.3EPYC 975478.3293.5340.1EPYC 9684X30.2301.3330.2Xeon Platinum 8490H65.7346.3357.7OpenBenchmarking.orgWatts, Fewer Is BetterGraph500 3.0CPU Power Consumption Monitor100200300400500

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: NUMAAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500SE +/- 2.51, N = 3SE +/- 1.10, N = 3SE +/- 4.03, N = 3SE +/- 2.90, N = 3SE +/- 1.38, N = 3SE +/- 1.16, N = 31454.681237.36967.561397.611173.071144.04-O2 -std=gnu99-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz-lm -laio -lapparmor -latomic -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz1. (CXX) g++ options: -lc
OpenBenchmarking.orgBogo Ops/s Per Watt, More Is BetterStress-NG 0.16.04Test: NUMAAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H51015202520.3694.0883.4504.9514.1063.603
MinAvgMaxAmpere Altra Max 128c22.371.481.9EPYC 965426.8302.7345.1EPYC 955419.9280.4320.9EPYC 975478.2282.3316.6EPYC 9684X30.0285.7325.7Xeon Platinum 8490H60.0317.5377.5OpenBenchmarking.orgWatts, Fewer Is BetterStress-NG 0.16.04CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 2452 Joules per run.2. EPYC 9654: Approximate power consumption of 10291 Joules per run.3. EPYC 9554: Approximate power consumption of 9628 Joules per run.4. EPYC 9754: Approximate power consumption of 9692 Joules per run.5. EPYC 9684X: Approximate power consumption of 9808 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 10796 Joules per run.

DuckDB

DuckDB is an in-progress SQL OLAP database management system optimized for analytics and features a vectorized and parallel engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDuckDB 0.9.1Benchmark: IMDBAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H306090120150SE +/- 0.18, N = 3SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 0.08, N = 3SE +/- 0.26, N = 3SE +/- 0.24, N = 3143.78115.81103.72147.60116.0599.251. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl
MinAvgMaxAmpere Altra Max 128c21.253.596.6EPYC 965427.0130.4182.9EPYC 955422.3112.3170.3EPYC 975476.8103.9150.1EPYC 9684X30.2142.7184.7Xeon Platinum 8490H55.4205.3262.2OpenBenchmarking.orgWatts, Fewer Is BetterDuckDB 0.9.1CPU Power Consumption Monitor701402102803501. Ampere Altra Max 128c: Approximate power consumption of 12280 Joules per run.2. EPYC 9654: Approximate power consumption of 20739 Joules per run.3. EPYC 9554: Approximate power consumption of 16094 Joules per run.4. EPYC 9754: Approximate power consumption of 20946 Joules per run.5. EPYC 9684X: Approximate power consumption of 22784 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 28264 Joules per run.

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 23.0.1Time To CompileAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H60120180240300SE +/- 3.13, N = 4SE +/- 1.09, N = 3SE +/- 1.67, N = 3SE +/- 2.85, N = 3SE +/- 1.97, N = 3SE +/- 1.53, N = 3260.33180.86179.74208.58180.06197.76
MinAvgMaxAmpere Altra Max 128c23.987.6173.4EPYC 965426.9166.9330.8EPYC 955421.7153.9325.7EPYC 975478.4144.8328.8EPYC 9684X29.3172.3326.1Xeon Platinum 8490H66.4228.2375.7OpenBenchmarking.orgWatts, Fewer Is BetterTimed Gem5 Compilation 23.0.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 23293 Joules per run.2. EPYC 9654: Approximate power consumption of 31092 Joules per run.3. EPYC 9554: Approximate power consumption of 28464 Joules per run.4. EPYC 9754: Approximate power consumption of 31026 Joules per run.5. EPYC 9684X: Approximate power consumption of 31932 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 46333 Joules per run.

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1530456075SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 366.0552.0549.9466.1352.9946.581. RawTherapee, version 5.9, command line.
MinAvgMaxAmpere Altra Max 128c22.154.578.6EPYC 965427.4122.5149.1EPYC 955422.2100.4123.5EPYC 975478.6106.6133.6EPYC 9684X30.6133.7159.4Xeon Platinum 8490H61.0187.2214.7OpenBenchmarking.orgWatts, Fewer Is BetterRawTherapeeCPU Power Consumption Monitor601201802403001. Ampere Altra Max 128c: Approximate power consumption of 3888 Joules per run.2. EPYC 9654: Approximate power consumption of 7024 Joules per run.3. EPYC 9554: Approximate power consumption of 5491 Joules per run.4. EPYC 9754: Approximate power consumption of 7601 Joules per run.5. EPYC 9684X: Approximate power consumption of 7756 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 9674 Joules per run.

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H300M600M900M1200M1500M987770000839917000865152000118451000088218100011131200001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200M400M600M800M1000M980463000816172000836159000114709000085402900010738600001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:10Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1.3M2.6M3.9M5.2M6.5MSE +/- 40896.93, N = 3SE +/- 18970.20, N = 3SE +/- 9013.12, N = 3SE +/- 45827.84, N = 3SE +/- 23729.89, N = 3SE +/- 14044.68, N = 35131209.296136092.716291456.535552637.525928614.584667128.011. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec Per Watt, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:10Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H8K16K24K32K40K39275.8523779.1128816.4125301.5923392.7415738.45
MinAvgMaxAmpere Altra Max 128c22.8130.6161.3EPYC 965426.4258.0300.7EPYC 955421.8218.3253.3EPYC 975478.2219.5245.8EPYC 9684X29.2253.4295.1Xeon Platinum 8490H63.8296.5336.2OpenBenchmarking.orgWatts, Fewer Is BetterMemcached 1.6.19CPU Power Consumption Monitor801602403204001. Ampere Altra Max 128c: Approximate power consumption of 9189 Joules per run.2. EPYC 9654: Approximate power consumption of 18063 Joules per run.3. EPYC 9554: Approximate power consumption of 15210 Joules per run.4. EPYC 9754: Approximate power consumption of 15435 Joules per run.5. EPYC 9684X: Approximate power consumption of 17741 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 20659 Joules per run.

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350SE +/- 0.75, N = 3SE +/- 0.41, N = 3SE +/- 0.61, N = 3SE +/- 2.48, N = 3SE +/- 0.50, N = 3SE +/- 0.60, N = 3307.88284.25238.29319.43283.87289.97
MinAvgMaxAmpere Altra Max 128c23.2159.6190.9EPYC 965426.6273.0321.3EPYC 955422.0274.1333.2EPYC 975478.2237.2310.5EPYC 9684X30.3269.2315.7Xeon Platinum 8490H64.6324.4376.4OpenBenchmarking.orgWatts, Fewer Is BetterTimed Linux Kernel Compilation 6.1CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 51561 Joules per run.2. EPYC 9654: Approximate power consumption of 81170 Joules per run.3. EPYC 9554: Approximate power consumption of 68798 Joules per run.4. EPYC 9754: Approximate power consumption of 79235 Joules per run.5. EPYC 9684X: Approximate power consumption of 79954 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 97739 Joules per run.

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H48121620SE +/- 0.21, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 316.7612.7913.7313.9914.7914.161. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
MinAvgMaxAmpere Altra Max 128c22.7118.0152.4EPYC 965427.1247.9302.4EPYC 955411.5208.0259.5EPYC 975479.5205.7244.3EPYC 9684X30.7237.3275.5Xeon Platinum 8490H65.3299.1351.3OpenBenchmarking.orgWatts, Fewer Is BetterPostgreSQL 16CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 17273 Joules per run.2. EPYC 9654: Approximate power consumption of 35289 Joules per run.3. EPYC 9554: Approximate power consumption of 29534 Joules per run.4. EPYC 9754: Approximate power consumption of 29481 Joules per run.5. EPYC 9684X: Approximate power consumption of 33703 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 42176 Joules per run.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H20K40K60K80K100KSE +/- 735.78, N = 3SE +/- 206.72, N = 3SE +/- 447.56, N = 3SE +/- 149.60, N = 3SE +/- 72.21, N = 3SE +/- 79.20, N = 35969378212728607148367606706041. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: EnhancedAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H30060090012001500SE +/- 2.73, N = 3SE +/- 0.88, N = 3SE +/- 0.88, N = 3SE +/- 0.58, N = 3SE +/- 1.67, N = 3SE +/- 0.33, N = 3123013191142145112771124-lxml2-lxml2-lxml2-lxml2-lxml21. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lzstd -lm -lpthread
OpenBenchmarking.orgIterations Per Minute Per Watt, More Is BetterGraphicsMagick 1.3.38Operation: EnhancedAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H36912159.8335.3234.8716.3285.1933.405
MinAvgMaxAmpere Altra Max 128c23.1125.1152.0EPYC 965426.3247.8266.4EPYC 955422.3234.4249.4EPYC 975480.1229.3244.6EPYC 9684X28.5245.9263.7Xeon Platinum 8490H57.6330.1364.4OpenBenchmarking.orgWatts, Fewer Is BetterGraphicsMagick 1.3.38CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 8005 Joules per run.2. EPYC 9654: Approximate power consumption of 15777 Joules per run.3. EPYC 9554: Approximate power consumption of 14926 Joules per run.4. EPYC 9754: Approximate power consumption of 14599 Joules per run.5. EPYC 9684X: Approximate power consumption of 15656 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 21128 Joules per run.

CPU Power Consumption Monitor

OpenBenchmarking.orgWattsCPU Power Consumption MonitorPhoronix Test Suite System MonitoringAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H150300450600750Min: 21.02 / Avg: 100.45 / Max: 241.47Min: 13.35 / Avg: 244.91 / Max: 809.65Min: 11.14 / Avg: 227.12 / Max: 369.31Min: 18.66 / Avg: 231.83 / Max: 866.74Min: 24.82 / Avg: 251.58 / Max: 451.13Min: 60.21 / Avg: 309.16 / Max: 380.81

Stockfish

MinAvgMaxAmpere Altra Max 128c23.1213.6244.8EPYC 965426.6320.1357.3EPYC 955422.7318.6349.2EPYC 975480.5322.3354.0EPYC 9684X31.0312.4349.6Xeon Platinum 8490H57.2334.5377.1OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption Monitor100200300400500

OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200K400K600K800K1000K881194.94902959.90758877.461127366.71902665.89458790.97

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total TimeAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H80M160M240M320M400MSE +/- 6034511.59, N = 15SE +/- 2088420.80, N = 3SE +/- 3637655.15, N = 15SE +/- 5395369.85, N = 15SE +/- 5369723.06, N = 15SE +/- 3106997.57, N = 15188202063289066909241774328363329642281982711153469265-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi21. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second Per Watt, More Is BetterStockfish 15Total TimeAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H200K400K600K800K1000K881194.94902959.90758877.461127366.71902665.89458790.97
MinAvgMaxAmpere Altra Max 128c23.1213.6244.8EPYC 965426.6320.1357.3EPYC 955422.7318.6349.2EPYC 975480.5322.3354.0EPYC 9684X31.0312.4349.6Xeon Platinum 8490H57.2334.5377.1OpenBenchmarking.orgWatts, Fewer Is BetterStockfish 15CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 16887 Joules per run.2. EPYC 9654: Approximate power consumption of 19208 Joules per run.3. EPYC 9554: Approximate power consumption of 19880 Joules per run.4. EPYC 9754: Approximate power consumption of 23849 Joules per run.5. EPYC 9684X: Approximate power consumption of 23367 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 29281 Joules per run.

CloverLeaf

MinAvgMaxAmpere Altra Max 128c23.3144.4161.6EPYC 965426.6214.7284.8EPYC 955421.8178.7248.8EPYC 975416.3174.2237.4EPYC 9684X28.7219.6291.4Xeon Platinum 8490H69.2322.0369.9OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeaf 1.3CPU Power Consumption Monitor100200300400500

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm64_shortAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H1428425670SE +/- 0.02, N = 3SE +/- 1.47, N = 12SE +/- 1.40, N = 12SE +/- 1.50, N = 12SE +/- 1.39, N = 12SE +/- 0.00, N = 362.7747.9845.8249.0351.3138.931. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
MinAvgMaxAmpere Altra Max 128c23.3144.4161.6EPYC 965426.6214.7284.8EPYC 955421.8178.7248.8EPYC 975416.3174.2237.4EPYC 9684X28.7219.6291.4Xeon Platinum 8490H69.2322.0369.9OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeaf 1.3CPU Power Consumption Monitor1002003004005001. Ampere Altra Max 128c: Approximate power consumption of 9869 Joules per run.2. EPYC 9654: Approximate power consumption of 11381 Joules per run.3. EPYC 9554: Approximate power consumption of 9101 Joules per run.4. EPYC 9754: Approximate power consumption of 9437 Joules per run.5. EPYC 9684X: Approximate power consumption of 12373 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 14167 Joules per run.

OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeaf 1.3CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 23.05 / Avg: 150.01 / Max: 164.12Min: 27.4 / Avg: 227.56 / Max: 288.41Min: 22.51 / Avg: 184.99 / Max: 244.7Min: 17.88 / Avg: 184.14 / Max: 237.7Min: 28.68 / Avg: 247.68 / Max: 306.65Min: 64.15 / Avg: 346.12 / Max: 376.87

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm16Ampere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H120240360480600SE +/- 0.15, N = 3SE +/- 15.29, N = 9SE +/- 4.26, N = 7SE +/- 14.10, N = 9SE +/- 14.65, N = 9SE +/- 0.02, N = 3550.30429.46460.24459.16298.29324.171. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgWatts, Fewer Is BetterCloverLeaf 1.3CPU Power Consumption MonitorAmpere Altra Max 128cEPYC 9654EPYC 9554EPYC 9754EPYC 9684XXeon Platinum 8490H70140210280350Min: 23.05 / Avg: 150.01 / Max: 164.12Min: 27.4 / Avg: 227.56 / Max: 288.41Min: 22.51 / Avg: 184.99 / Max: 244.7Min: 17.88 / Avg: 184.14 / Max: 237.7Min: 28.68 / Avg: 247.68 / Max: 306.65Min: 64.15 / Avg: 346.12 / Max: 376.871. Ampere Altra Max 128c: Approximate power consumption of 83054 Joules per run.2. EPYC 9654: Approximate power consumption of 98257 Joules per run.3. EPYC 9554: Approximate power consumption of 85597 Joules per run.4. EPYC 9754: Approximate power consumption of 85034 Joules per run.5. EPYC 9684X: Approximate power consumption of 74607 Joules per run.6. Xeon Platinum 8490H: Approximate power consumption of 113183 Joules per run.

91 Results Shown

NAS Parallel Benchmarks
John The Ripper
NAS Parallel Benchmarks
easyWave:
  e2Asean Grid + BengkuluSept2007 Source - 2400
  e2Asean Grid + BengkuluSept2007 Source - 1200
NAS Parallel Benchmarks
John The Ripper
Xmrig
Liquid-DSP
miniBUDE:
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
OpenSSL:
  ChaCha20-Poly1305
  ChaCha20
NAS Parallel Benchmarks
Liquid-DSP
libxsmm
NAS Parallel Benchmarks
ASKAP
Sysbench
libxsmm
OpenSSL
Xcompact3d Incompact3d
NAS Parallel Benchmarks
oneDNN
OpenSSL
ASKAP
Stress-NG
OpenSSL
7-Zip Compression
Stress-NG:
  Vector Math
  AVX-512 VNNI
ACES DGEMM
Primesieve
OpenSSL
Timed Linux Kernel Compilation
Timed Godot Game Engine Compilation
RocksDB
ClickHouse
Timed Node.js Compilation
Stress-NG
Algebraic Multi-Grid Benchmark
asmFish
miniFE
John The Ripper:
  bcrypt
  Blowfish
ClickHouse
Apache Cassandra
ClickHouse
PostgreSQL:
  100 - 1000 - Read Only
  100 - 1000 - Read Only - Average Latency
Coremark
Timed LLVM Compilation
Xmrig
ASKAP
NAS Parallel Benchmarks
Stress-NG
Memcached
GraphicsMagick
NWChem
Stress-NG
RocksDB
7-Zip Compression
Helsing
Stress-NG:
  Fused Multiply-Add
  Floating Point
Graph500
Xcompact3d Incompact3d
Rodinia
DuckDB
LULESH
High Performance Conjugate Gradient
Graph500
Stress-NG
DuckDB
Timed Gem5 Compilation
RawTherapee
Graph500:
  26:
    bfs max_TEPS
    bfs median_TEPS
Memcached
Timed Linux Kernel Compilation
PostgreSQL:
  100 - 1000 - Read Write - Average Latency
  100 - 1000 - Read Write
GraphicsMagick
CPU Power Consumption Monitor:
  Phoronix Test Suite System Monitoring
  CPU Power Consumption Monitor
  Total Time
Stockfish
CloverLeaf
CloverLeaf
CloverLeaf
CloverLeaf

Xeon Platinum 8490H

Processor: Intel Xeon Platinum 8490H @ 3.50GHz (60 Cores / 120 Threads), Motherboard: Quanta Cloud S6Q-MB-MPS (3A10.uh BIOS), Chipset: Intel Device 1bce, Memory: 512GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 31 October 2023 17:20 by user phoronix.

EPYC 9684X

Processor: AMD EPYC 9684X 96-Core @ 2.55GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 8 November 2023 16:47 by user phoronix.

EPYC 9754

Processor: AMD EPYC 9754 128-Core @ 2.25GHz (128 Cores / 256 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00116
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 9 November 2023 20:48 by user phoronix.

EPYC 9554

Processor: AMD EPYC 9554 64-Core @ 3.10GHz (64 Cores / 128 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 November 2023 02:51 by user phoronix.

EPYC 9654

Processor: AMD EPYC 9654 96-Core @ 2.40GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 15 November 2023 22:00 by user phoronix.

Ampere Altra Max 128c

Processor: ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores), Motherboard: GIGABYTE MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS), Chipset: Ampere Computing LLC Altra PCI Root Complex A, Memory: 512GB, Disk: 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Intel I350

OS: Ubuntu 23.10, Kernel: 6.6.0-rc5-phx-patched (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 27 November 2023 17:46 by user phoronix.