Benchmarks by Michael Larabel in the works for a future article on Phoronix.
EPYC 8324P Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 8324P - 155W EPYC 8324P - Power Processor: AMD EPYC 8324P 32-Core @ 2.65GHz (32 Cores / 64 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32 GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: llvmpipe, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7, OpenGL: 4.5 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 256 bits), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 640x480
EPYC 8324PN EPYC 8324PN - Power Processor: AMD EPYC 8534PN 32-Core @ 2.05GHz (32 Cores / 64 Threads)Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32 GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEEDNetwork: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 640x480
Xeon Gold 6421N Processor: Intel Xeon Gold 6421N @ 3.60GHz (32 Cores / 64 Threads)Motherboard: ASRockRack SPC741D8-2L2T/BCM (1.26 BIOS)Chipset: Intel Device 1bceMemory: 8 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: 2 x Intel I210 + 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (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: 640x480
Kernel Notes: Transparent Huge Pages: madviseCompiler 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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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
Xeon Gold 6421N - 6c RAM Changed Memory to 6 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1 .
AMD EPYC 8324P 8324PN Siena Benchmarks Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution EPYC 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM AMD EPYC 8324P 32-Core @ 2.65GHz (32 Cores / 64 Threads) AMD Cinnabar (RCB1009C BIOS) AMD Device 14a4 6 x 32 GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS llvmpipe 2 x Broadcom NetXtreme BCM5720 PCIe Ubuntu 23.10 6.5.0-5-generic (x86_64) GNOME Shell X Server 1.21.1.7 4.5 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 256 bits) GCC 13.2.0 ext4 640x480 AMD EPYC 8534PN 32-Core @ 2.05GHz (32 Cores / 64 Threads) ASPEED Intel Xeon Gold 6421N @ 3.60GHz (32 Cores / 64 Threads) ASRockRack SPC741D8-2L2T/BCM (1.26 BIOS) Intel Device 1bce 8 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1 2 x Intel I210 + 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA GNOME Shell 45.0 6 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - EPYC 8324P: --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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 8324P - 155W: --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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 8324P - Power: --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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 8324PN: --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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 8324PN - Power: --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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 - Xeon Gold 6421N: --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 - Xeon Gold 6421N - 6c RAM: --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 Disk Details - NONE / relatime,rw,stripe=64 / Block Size: 4096 Processor Details - EPYC 8324P: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 - EPYC 8324P - 155W: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 - EPYC 8324P - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 - EPYC 8324PN: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 - EPYC 8324PN - Power: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212 - Xeon Gold 6421N: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1 - Xeon Gold 6421N - 6c RAM: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1 Java Details - OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1) Python Details - Python 3.11.5 Security Details - EPYC 8324P: 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 8324P - 155W: 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 8324P - Power: 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 8324PN: 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 8324PN - Power: 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 - Xeon Gold 6421N: 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 - Xeon Gold 6421N - 6c RAM: 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 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM Result Overview Phoronix Test Suite 100% 117% 134% 152% 169% Sysbench YugabyteDB ebizzy Memcached CockroachDB Apache Cassandra TensorFlow RocksDB LAMMPS Molecular Dynamics Simulator Blender Chaos Group V-RAY TiDB Community Server 7-Zip Compression Timed Linux Kernel Compilation GPAW OpenSSL LuxCoreRender ClickHouse OpenRadioss Appleseed Timed LLVM Compilation Liquid-DSP ASTC Encoder Kvazaar Timed Node.js Compilation Timed Godot Game Engine Compilation uvg266 SVT-AV1 John The Ripper Apache IoTDB
EPYC 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM Per Watt Result Overview Phoronix Test Suite 100% 138% 177% 215% 254% Sysbench 7-Zip Compression Apache Cassandra RocksDB uvg266 Kvazaar YugabyteDB Liquid-DSP LuxCoreRender John The Ripper SVT-AV1 ebizzy Chaos Group V-RAY ASTC Encoder OpenSSL ClickHouse LAMMPS Molecular Dynamics Simulator Memcached TensorFlow TiDB Community Server CockroachDB P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M P.W.G.M
AMD EPYC 8324P 8324PN Siena Benchmarks openssl: AES-128-GCM sysbench: CPU memcached: 1:10 openssl: RSA4096 openssl: AES-256-GCM ebizzy: cockroach: KV, 50% Reads - 128 cassandra: Writes compress-7zip: Decompression Rating memcached: 1:100 openssl: SHA256 tensorflow: CPU - 64 - ResNet-50 openradioss: Chrysler Neon 1M tensorflow: CPU - 32 - ResNet-50 rocksdb: Rand Read tidb: oltp_update_non_index - 128 blender: Pabellon Barcelona - CPU-Only lammps: 20k Atoms blender: Fishy Cat - CPU-Only blender: BMW27 - CPU-Only blender: Classroom - CPU-Only tensorflow: CPU - 16 - ResNet-50 v-ray: CPU blender: Barbershop - CPU-Only appleseed: Disney Material luxcorerender: DLSC - CPU build-linux-kernel: allmodconfig luxcorerender: LuxCore Benchmark - CPU gpaw: Carbon Nanotube openssl: SHA512 luxcorerender: Danish Mood - CPU john-the-ripper: MD5 luxcorerender: Orange Juice - CPU kvazaar: Bosphorus 4K - Medium astcenc: Exhaustive john-the-ripper: Blowfish john-the-ripper: bcrypt clickhouse: 100M Rows Hits Dataset, Third Run uvg266: Bosphorus 4K - Medium clickhouse: 100M Rows Hits Dataset, Second Run clickhouse: 100M Rows Hits Dataset, First Run / Cold Cache kvazaar: Bosphorus 4K - Very Fast build-llvm: Ninja uvg266: Bosphorus 4K - Super Fast tidb: oltp_update_non_index - 64 uvg266: Bosphorus 1080p - Medium liquid-dsp: 64 - 256 - 32 liquid-dsp: 128 - 256 - 32 uvg266: Bosphorus 4K - Very Fast astcenc: Thorough kvazaar: Bosphorus 4K - Super Fast openradioss: INIVOL and Fluid Structure Interaction Drop Container build-nodejs: Time To Compile appleseed: Emily build-godot: Time To Compile uvg266: Bosphorus 4K - Ultra Fast luxcorerender: Rainbow Colors and Prism - CPU uvg266: Bosphorus 1080p - Very Fast apache-iotdb: 500 - 100 - 800 - 100 svt-av1: Preset 8 - Bosphorus 4K uvg266: Bosphorus 1080p - Super Fast john-the-ripper: HMAC-SHA512 compress-7zip: Compression Rating astcenc: Medium openssl: ChaCha20-Poly1305 apache-iotdb: 800 - 100 - 500 - 100 apache-iotdb: 500 - 100 - 800 - 100 apache-iotdb: 800 - 100 - 500 - 100 kvazaar: Bosphorus 4K - Ultra Fast uvg266: Bosphorus 1080p - Ultra Fast apache-iotdb: 800 - 100 - 800 - 100 apache-iotdb: 800 - 100 - 800 - 100 john-the-ripper: WPA PSK openssl: ChaCha20 yugabytedb: CassandraBatchKeyValue, Batch 25 - 32 - 32 EPYC 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM 305504550637 124680.00 3770609.67 14926.2 263139293050 1891396 84158.3 226400 211252 3753320.91 43014953050 67.97 384.47 61.57 160510214 46561 146.67 18.312 58.96 47.02 118.08 51.60 34425 429.18 75.624094 5.54 463.482 4.56 81.671 13803364873 4.22 5251333 8.66 20.14 2.1194 55462 55494 360.01 15.60 349.27 334.83 41.92 258.133 42.56 36397 43.95 1710466667 1728433333 38.79 20.3996 53.91 222.82 223.341 152.811359 163.317 46.20 19.39 120.39 75.20 78.699 136.57 170626333 239650 160.7468 113098144337 50.54 98984707 93807432 62.18 148.70 72.83 104847158 203025 159441828337 108862.05 278325833270 124675.31 3587265.42 13423.1 240586561853 1717958 79433.6 224054 198073 3549968.36 39868583800 63.83 387.08 57.95 145926007 45300 160.30 17.022 64.42 51.58 128.85 48.73 31510 465.53 81.779192 5.13 499.360 4.21 84.909 12889684360 3.91 4831667 8.11 19.19 1.9475 53523 53549 347.66 14.49 341.20 329.38 40.74 275.469 40.32 36088 44.07 1603333333 1612933333 37.06 19.3379 51.82 232.65 235.681 160.880587 171.392 44.70 18.60 120.19 77.84 74.621 134.92 171710333 228756 154.1378 111478070750 50.97 94808492 93050746 61.09 148.95 75.43 101322673 202843 159437074800 75678.99 315571591120 124677.13 3745312.12 18742.4 269992654257 2316454 84204.1 220692 212484 3755206.01 43047518260 68.88 385.28 61.89 175960103 46225 135.27 18.642 54.34 43.69 109.56 51.61 36903 404.84 71.607641 5.87 438.700 4.80 80.173 13812459643 4.43 5385000 9.13 20.30 2.1608 55520 55469 359.88 16.00 354.66 341.29 42.00 250.503 43.73 36662 44.13 1713766667 1729733333 39.88 20.6131 54.10 219.55 218.726 149.430439 158.066 47.35 19.57 120.82 75.40 80.051 136.34 171768000 240409 161.2220 113108043093 49.20 97619880 96383625 61.93 149.14 73.05 104603173 202775 159433379203 109894.39 250022699553 123969.55 3319924.38 11771.3 214405423833 1525748 73141.1 209500 177029 3255564.05 35571083230 60.02 396.48 54.51 129169541 41185 180.89 15.257 72.32 58.10 145.43 47.10 28044 524.37 91.924597 4.57 556.671 3.79 91.179 11578262873 3.52 4291333 7.29 17.32 1.7295 47990 48143 332.63 12.95 332.04 321.56 37.11 301.300 36.40 33484 42.43 1432866667 1446866667 33.41 17.3111 47.11 255.09 255.463 174.063532 184.529 40.68 17.13 113.93 83.67 69.458 127.09 163350000 211839 141.9828 99907867353 53.16 89411612 89550933 56.06 141.61 79.54 96286895 183737 146313390400 70619.32 314757015327 124680.87 3791375.58 18163.4 270018374310 2241998 84727.0 221420 212569 3790339.96 43034165953 68.40 384.95 61.78 176583360 47223 135.64 18.620 54.66 43.70 109.88 51.59 36584 407.87 71.544415 5.75 441.439 4.80 80.790 13803852343 4.34 5335333 9.03 20.30 2.1580 55526 55546 355.02 15.96 357.13 343.11 42.01 252.311 43.57 36829 43.96 1713066667 1726900000 39.63 20.5647 54.05 219.44 218.933 149.033455 159.060 47.08 19.90 121.09 76.76 79.377 137.07 171762667 241268 161.4424 113119629900 49.99 97338332 94636861 62.09 149.56 72.44 105469536 203093 159458412410 110313.78 426543281130 73882.86 5297302.81 14790.1 335006361980 1874474 108299.0 157222 145842 4634802.76 30387621850 84.87 307.10 75.61 135717334 55478 161.81 20.319 64.68 47.78 129.90 62.49 32282 497.81 75.139129 5.10 480.689 4.50 72.126 10969551063 4.15 5284667 8.38 16.24 2.0103 46376 46374 395.62 13.31 390.25 373.19 34.94 264.194 37.35 40075 36.99 1575300000 1593866667 33.82 20.5434 45.92 216.87 221.950 154.279595 162.846 42.53 19.71 104.53 72.36 74.693 119.20 186859333 227218 150.7795 107133667010 47.10 100840933 100615762 55.60 134.08 71.50 106918573 192699 149212297717 96623.75 426186033560 73881.56 5341589.48 14873.9 333680592760 1901321 107019.0 153009 145846 4635050.16 30346438850 70.94 433.93 66.81 145259860 52850 161.39 20.248 64.78 47.74 129.66 58.25 32511 496.97 75.165385 5.11 482.265 4.50 86.309 10951012703 4.15 5016667 8.39 16.21 2.0103 44577 44724 319.17 13.28 317.23 303.78 34.84 268.279 37.36 39692 36.89 1575233333 1590300000 33.74 20.5415 45.85 228.84 224.702 154.411491 164.094 42.48 19.79 104.57 80.16 74.501 119.12 174776333 218723 150.8148 106988442053 51.99 92766907 91362622 55.45 134.18 75.64 101011034 185188 149077028223 106397.44 OpenBenchmarking.org
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.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 90000M 180000M 270000M 360000M 450000M SE +/- 292984177.38, N = 3 SE +/- 914543910.24, N = 3 SE +/- 390762020.13, N = 3 SE +/- 437719400.62, N = 3 SE +/- 89134512.42, N = 3 SE +/- 369408212.75, N = 3 SE +/- 320549083.82, N = 3 250022699553 278325833270 305504550637 314757015327 315571591120 426186033560 426543281130 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 600M 1200M 1800M 2400M 3000M 2272511680.16 2407233038.50 2428236149.06 2561020004.81 2574143857.10 2684165407.73 2896109007.34 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 60.8 177.0 186.9 Xeon Gold 6421N 67.9 175.7 192.5 EPYC 8324P - Power 19.3 138.9 148.3 EPYC 8324PN - Power 17.5 122.3 134.4 EPYC 8324P 18.8 119.3 125.2 EPYC 8324P - 155W 18.2 103.7 109.2 EPYC 8324PN 17.5 86.3 91.4 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power 30K 60K 90K 120K 150K SE +/- 1.06, N = 3 SE +/- 0.99, N = 3 SE +/- 45.55, N = 3 SE +/- 3.90, N = 3 SE +/- 5.03, N = 3 SE +/- 6.97, N = 3 SE +/- 14.40, N = 3 73881.56 73882.86 123969.55 124675.31 124677.13 124680.00 124680.87 1. (CC) gcc options: -O2 -funroll-loops -rdynamic -ldl -laio -lm Events Per Second Per Watt
OpenBenchmarking.org Events Per Second Per Watt, More Is Better Sysbench 1.0.20 Test: CPU Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 400 800 1200 1600 2000 650.52 655.91 1347.57 1358.09 1358.95 1421.61 1648.86 Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 63.2 113.6 117.7 Xeon Gold 6421N 56.0 112.6 119.3 EPYC 8324P 28.9 92.5 96.8 EPYC 8324P - Power 16.5 91.8 96.8 EPYC 8324P - 155W 25.9 91.7 96.8 EPYC 8324PN - Power 18.2 87.7 92.4 EPYC 8324PN 18.9 75.2 79.8 OpenBenchmarking.org Watts, Fewer Is Better Sysbench 1.0.20 CPU Power Consumption Monitor 40 80 120 160 200
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.
Result
OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM 1.1M 2.2M 3.3M 4.4M 5.5M SE +/- 4227.64, N = 3 SE +/- 9343.47, N = 3 SE +/- 3548.13, N = 3 SE +/- 8759.20, N = 3 SE +/- 11712.03, N = 3 SE +/- 56555.57, N = 3 SE +/- 59196.74, N = 15 3319924.38 3587265.42 3745312.12 3770609.67 3791375.58 5297302.81 5341589.48 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre Ops/sec Per Watt
OpenBenchmarking.org Ops/sec Per Watt, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 10K 20K 30K 40K 50K 33811.39 33815.06 36211.64 36609.50 38728.71 40686.91 45803.91 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.2 158.0 178.0 Xeon Gold 6421N 57.4 156.7 177.0 EPYC 8324P 19.8 104.1 120.7 EPYC 8324P - Power 18.9 102.3 120.4 EPYC 8324PN - Power 17.5 97.9 113.8 EPYC 8324P - 155W 19.4 88.2 101.3 EPYC 8324PN 16.7 72.5 84.5 OpenBenchmarking.org Watts, Fewer Is Better Memcached 1.6.19 CPU Power Consumption Monitor 50 100 150 200 250
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.org sign/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 4K 8K 12K 16K 20K SE +/- 39.06, N = 3 SE +/- 7.33, N = 3 SE +/- 31.24, N = 3 SE +/- 48.13, N = 3 SE +/- 7.48, N = 3 SE +/- 11.63, N = 3 SE +/- 12.31, N = 3 11771.3 13423.1 14790.1 14873.9 14926.2 18163.4 18742.4 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 70000M 140000M 210000M 280000M 350000M SE +/- 573541074.64, N = 3 SE +/- 127258426.38, N = 3 SE +/- 59738149.52, N = 3 SE +/- 65175324.71, N = 3 SE +/- 15608226.68, N = 3 SE +/- 1103616965.79, N = 3 SE +/- 861378679.86, N = 3 214405423833 240586561853 263139293050 269992654257 270018374310 333680592760 335006361980 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 500M 1000M 1500M 2000M 2500M 1916133269.71 1932173717.91 1988020192.98 2116936172.47 2211314784.44 2317422384.43 2475107862.54 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.4 174.1 190.9 Xeon Gold 6421N 62.3 173.4 190.8 EPYC 8324P - Power 19.4 135.8 145.2 EPYC 8324PN - Power 34.8 127.6 134.4 EPYC 8324P 19.3 119.0 125.8 EPYC 8324P - 155W 33.3 103.8 107.8 EPYC 8324PN 17.8 86.6 91.1 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
ebizzy This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N EPYC 8324P Xeon Gold 6421N - 6c RAM EPYC 8324PN - Power EPYC 8324P - Power 500K 1000K 1500K 2000K 2500K SE +/- 15264.73, N = 3 SE +/- 13723.58, N = 9 SE +/- 9111.58, N = 3 SE +/- 16202.22, N = 3 SE +/- 22744.33, N = 15 SE +/- 15496.66, N = 3 SE +/- 15806.66, N = 3 1525748 1717958 1874474 1891396 1901321 2241998 2316454 1. (CC) gcc options: -pthread -lpthread -O3 -march=native Records/s Per Watt
OpenBenchmarking.org Records/s Per Watt, More Is Better ebizzy 0.3 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN EPYC 8324P 5K 10K 15K 20K 25K 11847.65 11862.47 18102.15 20750.27 20810.58 21268.80 21642.54 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 61.6 160.5 186.9 Xeon Gold 6421N 56.6 158.0 185.1 EPYC 8324P - Power 20.1 128.0 158.8 EPYC 8324PN - Power 18.3 107.7 134.6 EPYC 8324P 16.2 87.4 111.8 EPYC 8324P - 155W 17.4 82.8 98.8 EPYC 8324PN 17.7 71.7 85.9 OpenBenchmarking.org Watts, Fewer Is Better ebizzy 0.3 CPU Power Consumption Monitor 50 100 150 200 250
CockroachDB CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 128 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 20K 40K 60K 80K 100K SE +/- 164.84, N = 3 SE +/- 265.96, N = 3 SE +/- 367.80, N = 3 SE +/- 160.15, N = 3 SE +/- 473.57, N = 3 SE +/- 110.69, N = 3 SE +/- 290.04, N = 3 73141.1 79433.6 84158.3 84204.1 84727.0 107019.0 108299.0 ops/s Per Watt
OpenBenchmarking.org ops/s Per Watt, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 128 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 200 400 600 800 1000 686.22 699.06 784.43 789.24 831.25 864.46 928.96 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 63.5 156.0 181.4 Xeon Gold 6421N 56.6 154.9 182.4 EPYC 8324P 17.1 107.3 125.2 EPYC 8324P - Power 18.4 106.7 125.4 EPYC 8324PN - Power 20.9 101.9 119.2 EPYC 8324P - 155W 18.7 91.9 106.0 EPYC 8324PN 20.3 78.7 85.8 OpenBenchmarking.org Watts, Fewer Is Better CockroachDB 22.2 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 4.1.3 Test: Writes Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324P 50K 100K 150K 200K 250K SE +/- 471.91, N = 3 SE +/- 355.83, N = 3 SE +/- 674.10, N = 3 SE +/- 1000.49, N = 3 SE +/- 378.26, N = 3 SE +/- 487.84, N = 3 SE +/- 527.41, N = 3 153009 157222 209500 220692 221420 224054 226400 Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better Apache Cassandra 4.1.3 Test: Writes Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324PN 600 1200 1800 2400 3000 1196.50 1230.05 2312.76 2329.94 2342.66 2397.15 2749.61 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 61.9 127.9 139.8 Xeon Gold 6421N 57.6 127.8 141.0 EPYC 8324P 27.4 96.6 108.6 EPYC 8324P - 155W 37.0 96.2 106.9 EPYC 8324P - Power 20.3 95.4 107.1 EPYC 8324PN - Power 31.6 92.4 101.7 EPYC 8324PN 21.6 76.2 84.5 OpenBenchmarking.org Watts, Fewer Is Better Apache Cassandra 4.1.3 CPU Power Consumption Monitor 40 80 120 160 200
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.
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 50K 100K 150K 200K 250K SE +/- 8.21, N = 3 SE +/- 16.60, N = 3 SE +/- 176.42, N = 3 SE +/- 382.80, N = 3 SE +/- 78.71, N = 3 SE +/- 209.37, N = 3 SE +/- 120.18, N = 3 145842 145846 177029 198073 211252 212484 212569 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC MIPS Per Watt
OpenBenchmarking.org MIPS Per Watt, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 500 1000 1500 2000 2500 931.79 939.01 2023.74 2094.16 2152.08 2320.49 2321.86 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.8 156.5 183.1 Xeon Gold 6421N - 6c RAM 55.1 155.3 178.4 EPYC 8324P - Power 19.8 105.0 129.6 EPYC 8324P 23.5 100.9 121.5 EPYC 8324PN - Power 17.5 98.8 123.4 EPYC 8324P - 155W 18.4 85.4 104.2 EPYC 8324PN 18.0 76.2 89.0 OpenBenchmarking.org Watts, Fewer Is Better 7-Zip Compression 22.01 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM 1000K 2000K 3000K 4000K 5000K SE +/- 7889.45, N = 3 SE +/- 12557.13, N = 3 SE +/- 4177.23, N = 3 SE +/- 8946.59, N = 3 SE +/- 13082.02, N = 3 SE +/- 48265.78, N = 15 SE +/- 71497.36, N = 15 3255564.05 3549968.36 3753320.91 3755206.01 3790339.96 4634802.76 4635050.16 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre Ops/sec Per Watt
OpenBenchmarking.org Ops/sec Per Watt, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 10K 20K 30K 40K 50K 28919.65 29000.25 35658.02 36553.39 38432.12 41375.16 44877.21 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 60.2 160.3 185.3 Xeon Gold 6421N 55.3 159.8 184.4 EPYC 8324P 18.2 105.3 123.4 EPYC 8324P - Power 18.3 102.7 122.9 EPYC 8324PN - Power 17.3 98.6 116.5 EPYC 8324P - 155W 17.6 85.8 101.0 EPYC 8324PN 17.0 72.5 84.4 OpenBenchmarking.org Watts, Fewer Is Better Memcached 1.6.19 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA256 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 9000M 18000M 27000M 36000M 45000M SE +/- 7186957.97, N = 3 SE +/- 1733431.16, N = 3 SE +/- 34178909.69, N = 3 SE +/- 20174746.76, N = 3 SE +/- 5819413.04, N = 3 SE +/- 6141870.83, N = 3 SE +/- 1365376.75, N = 3 30346438850 30387621850 35571083230 39868583800 43014953050 43034165953 43047518260 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: SHA256 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 90M 180M 270M 360M 450M 188388995.40 190068851.91 342898823.16 375769686.67 382152888.04 405880062.10 434931611.82 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 84.1 161.1 186.6 Xeon Gold 6421N 56.6 159.9 186.7 EPYC 8324P - Power 19.5 125.5 148.9 EPYC 8324P 22.4 114.5 122.0 EPYC 8324PN - Power 39.7 112.6 134.5 EPYC 8324P - 155W 19.1 98.2 104.2 EPYC 8324PN 17.4 81.8 87.7 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 60.02 63.83 67.97 68.40 68.88 70.94 84.87 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.1505 0.301 0.4515 0.602 0.7525 0.464 0.535 0.575 0.579 0.604 0.623 0.669 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.2 158.8 170.1 Xeon Gold 6421N - 6c RAM 58.7 152.9 162.5 EPYC 8324P - Power 19.3 119.7 130.1 EPYC 8324P 19.0 117.4 127.4 EPYC 8324PN - Power 17.4 113.3 121.9 EPYC 8324P - 155W 20.6 102.4 109.1 EPYC 8324PN 17.3 89.7 96.4 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 50 100 150 200 250
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 20 40 60 80 100 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 54.51 57.95 61.57 61.78 61.89 66.81 75.61 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 0.1519 0.3038 0.4557 0.6076 0.7595 0.468 0.516 0.561 0.565 0.592 0.593 0.675 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 62.8 146.6 161.8 Xeon Gold 6421N - 6c RAM 60.3 142.8 157.0 EPYC 8324P 18.8 109.7 123.1 EPYC 8324P - Power 18.6 109.5 124.8 EPYC 8324PN - Power 17.0 104.2 118.3 EPYC 8324P - 155W 24.7 97.9 107.7 EPYC 8324PN 17.3 80.7 88.9 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Random Read EPYC 8324PN Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 40M 80M 120M 160M 200M SE +/- 408101.80, N = 3 SE +/- 3107107.33, N = 15 SE +/- 2002076.54, N = 12 SE +/- 99959.05, N = 3 SE +/- 93851.87, N = 3 SE +/- 1054694.33, N = 3 SE +/- 122745.09, N = 3 129169541 135717334 145259860 145926007 160510214 175960103 176583360 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread Op/s Per Watt
OpenBenchmarking.org Op/s Per Watt, More Is Better RocksDB 8.0 Test: Random Read Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 400K 800K 1200K 1600K 2000K 770740.13 824200.80 1337631.19 1416799.34 1552072.98 1610282.84 1694284.05 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 56.9 176.2 189.5 Xeon Gold 6421N 55.4 176.1 189.1 EPYC 8324P - Power 18.5 131.5 144.0 EPYC 8324PN - Power 17.0 124.6 134.2 EPYC 8324P 33.5 103.4 110.7 EPYC 8324P - 155W 19.1 90.6 97.5 EPYC 8324PN 17.9 76.2 83.3 OpenBenchmarking.org Watts, Fewer Is Better RocksDB 8.0 CPU Power Consumption Monitor 50 100 150 200 250
TiDB Community Server This is a PingCAP TiDB Community Server benchmark facilitated using the sysbench OLTP database benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Queries Per Second, More Is Better TiDB Community Server 7.3 Test: oltp_update_non_index - Threads: 128 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 12K 24K 36K 48K 60K SE +/- 110.11, N = 3 SE +/- 22.86, N = 3 SE +/- 128.67, N = 3 SE +/- 127.93, N = 3 SE +/- 194.94, N = 3 SE +/- 1224.64, N = 9 SE +/- 168.43, N = 3 41185 45300 46225 46561 47223 52850 55478 Queries Per Second Per Watt
OpenBenchmarking.org Queries Per Second Per Watt, More Is Better TiDB Community Server 7.3 Test: oltp_update_non_index - Threads: 128 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 130 260 390 520 650 447.62 462.75 556.25 556.42 583.53 585.75 597.28 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 56.6 119.9 165.0 Xeon Gold 6421N - 6c RAM 58.3 118.1 166.9 EPYC 8324P 17.8 83.7 117.6 EPYC 8324P - Power 20.2 83.1 116.9 EPYC 8324PN - Power 17.2 80.6 111.5 EPYC 8324P - 155W 18.3 77.6 107.9 EPYC 8324PN 17.5 69.0 83.4 OpenBenchmarking.org Watts, Fewer Is Better TiDB Community Server 7.3 CPU Power Consumption Monitor 50 100 150 200 250
LAMMPS Molecular Dynamics Simulator LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 15.26 17.02 18.31 18.62 18.64 20.25 20.32 1. (CXX) g++ options: -O3 -lm -ldl ns/day Per Watt
OpenBenchmarking.org ns/day Per Watt, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.041 0.082 0.123 0.164 0.205 0.111 0.112 0.147 0.155 0.163 0.173 0.182 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.7 182.1 184.9 Xeon Gold 6421N - 6c RAM 56.7 181.9 184.7 EPYC 8324P - Power 19.4 126.6 130.6 EPYC 8324PN - Power 17.7 120.0 121.9 EPYC 8324P 21.2 112.7 116.8 EPYC 8324P - 155W 19.0 98.6 100.3 EPYC 8324PN 17.6 83.8 85.2 OpenBenchmarking.org Watts, Fewer Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 CPU Power Consumption Monitor 50 100 150 200 250
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 EPYC 8324PN EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324P EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 14 28 42 56 70 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 47.10 48.73 51.59 51.60 51.61 58.25 62.49 images/sec Per Watt
OpenBenchmarking.org images/sec Per Watt, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN EPYC 8324P - 155W EPYC 8324PN - Power 0.1242 0.2484 0.3726 0.4968 0.621 0.437 0.477 0.528 0.533 0.547 0.549 0.552 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 63.9 133.3 150.7 Xeon Gold 6421N 55.3 130.9 153.4 EPYC 8324P - Power 19.4 97.7 116.5 EPYC 8324P 18.1 96.9 115.8 EPYC 8324PN - Power 18.0 93.5 111.1 EPYC 8324P - 155W 18.2 88.8 104.4 EPYC 8324PN 24.3 86.1 97.9 OpenBenchmarking.org Watts, Fewer Is Better TensorFlow 2.12 CPU Power Consumption Monitor 40 80 120 160 200
Chaos Group V-RAY This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org vsamples, More Is Better Chaos Group V-RAY 5.02 Mode: CPU EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 8K 16K 24K 32K 40K SE +/- 56.22, N = 3 SE +/- 114.78, N = 3 SE +/- 101.37, N = 3 SE +/- 30.68, N = 3 SE +/- 64.16, N = 3 SE +/- 79.43, N = 3 SE +/- 71.00, N = 3 28044 31510 32282 32511 34425 36584 36903 vsamples Per Watt
OpenBenchmarking.org vsamples Per Watt, More Is Better Chaos Group V-RAY 5.02 Mode: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 80 160 240 320 400 206.92 207.44 317.38 349.82 356.73 361.66 369.03 Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.1 156.7 188.4 Xeon Gold 6421N 55.8 156.0 186.6 EPYC 8324P - Power 18.9 116.3 149.6 EPYC 8324PN - Power 17.6 101.2 127.2 EPYC 8324P 19.1 98.4 118.9 EPYC 8324P - 155W 20.2 88.3 107.7 EPYC 8324PN 17.4 76.0 90.3 OpenBenchmarking.org Watts, Fewer Is Better Chaos Group V-RAY 5.02 CPU Power Consumption Monitor 50 100 150 200 250
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 8324PN Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 1.3208 2.6416 3.9624 5.2832 6.604 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 4.57 5.10 5.11 5.13 5.54 5.75 5.87 MIN: 4.42 / MAX: 4.94 MIN: 4.95 / MAX: 5.5 MIN: 4.96 / MAX: 5.52 MIN: 4.97 / MAX: 5.53 MIN: 5.39 / MAX: 6.02 MIN: 5.62 / MAX: 6.2 MIN: 5.7 / MAX: 6.29 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.0126 0.0252 0.0378 0.0504 0.063 0.029 0.030 0.045 0.048 0.052 0.055 0.056 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 58.3 174.3 187.2 Xeon Gold 6421N - 6c RAM 56.1 172.9 189.3 EPYC 8324P - Power 21.8 131.5 144.3 EPYC 8324PN - Power 18.4 120.4 132.1 EPYC 8324P 20.6 107.1 117.7 EPYC 8324P - 155W 19.2 93.8 104.3 EPYC 8324PN 21.3 81.9 89.4 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250
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.
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 1.08 2.16 3.24 4.32 5.4 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.79 4.21 4.50 4.50 4.56 4.80 4.80 MIN: 1.35 / MAX: 4.5 MIN: 1.54 / MAX: 4.97 MIN: 1.81 / MAX: 5.22 MIN: 1.8 / MAX: 5.23 MIN: 1.69 / MAX: 5.35 MIN: 1.77 / MAX: 5.65 MIN: 1.76 / MAX: 5.64 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: LuxCore Benchmark - Acceleration: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.0106 0.0212 0.0318 0.0424 0.053 0.026 0.026 0.039 0.041 0.041 0.043 0.047 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 68.7 173.1 187.3 Xeon Gold 6421N 61.0 170.2 187.2 EPYC 8324P - Power 18.9 121.7 139.3 EPYC 8324PN - Power 17.5 115.9 132.0 EPYC 8324P 20.5 111.8 125.9 EPYC 8324P - 155W 18.8 97.4 108.7 EPYC 8324PN 17.5 81.0 89.3 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA512 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 3000M 6000M 9000M 12000M 15000M SE +/- 3067599.31, N = 3 SE +/- 1113058.97, N = 3 SE +/- 807446.47, N = 3 SE +/- 915924.04, N = 3 SE +/- 1359364.00, N = 3 SE +/- 789257.32, N = 3 SE +/- 5078869.31, N = 3 10951012703 10969551063 11578262873 12889684360 13803364873 13803852343 13812459643 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: SHA512 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 30M 60M 90M 120M 150M 63439216.78 63877243.16 109614321.25 116352379.16 124111653.80 129372673.92 141241147.58 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 60.7 172.6 185.6 Xeon Gold 6421N 55.8 171.7 184.2 EPYC 8324P - Power 20.9 126.0 140.8 EPYC 8324PN - Power 17.9 118.6 132.1 EPYC 8324P 18.8 111.2 120.9 EPYC 8324P - 155W 19.0 99.6 106.3 EPYC 8324PN 18.1 82.0 87.5 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 0.9968 1.9936 2.9904 3.9872 4.984 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 3.52 3.91 4.15 4.15 4.22 4.34 4.43 MIN: 1.34 / MAX: 4.12 MIN: 1.56 / MAX: 4.59 MIN: 1.63 / MAX: 4.82 MIN: 1.68 / MAX: 4.83 MIN: 1.74 / MAX: 4.93 MIN: 1.77 / MAX: 5.06 MIN: 1.82 / MAX: 5.14 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.0097 0.0194 0.0291 0.0388 0.0485 0.024 0.024 0.036 0.037 0.038 0.041 0.043 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.2 172.0 186.7 Xeon Gold 6421N 56.1 171.4 187.1 EPYC 8324P - Power 19.5 121.0 137.0 EPYC 8324P 19.5 115.7 128.4 EPYC 8324PN - Power 17.2 115.5 129.0 EPYC 8324P - 155W 18.6 94.9 108.5 EPYC 8324PN 17.5 81.4 89.5 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: MD5 EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N - 6c RAM EPYC 8324P Xeon Gold 6421N EPYC 8324PN - Power EPYC 8324P - Power 1.2M 2.4M 3.6M 4.8M 6M SE +/- 4807.40, N = 3 SE +/- 5364.49, N = 3 SE +/- 8875.68, N = 3 SE +/- 1452.97, N = 3 SE +/- 3382.96, N = 3 SE +/- 3480.10, N = 3 SE +/- 4618.80, N = 3 4291333 4831667 5016667 5251333 5284667 5335333 5385000 -lbz2 -lbz2 -lbz2 -lbz2 -lbz2 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: MD5 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 11K 22K 33K 44K 55K 28929.38 30496.92 41878.83 43806.10 44865.89 48352.81 50632.61 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 61.4 173.4 191.6 Xeon Gold 6421N 56.6 173.3 187.2 EPYC 8324P - Power 19.9 128.6 147.2 EPYC 8324P 15.8 119.9 133.0 EPYC 8324PN - Power 17.2 118.9 134.5 EPYC 8324P - 155W 19.5 99.9 111.1 EPYC 8324PN 23.5 84.8 92.9 OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor 50 100 150 200 250
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 7.29 8.11 8.38 8.39 8.66 9.03 9.13 MIN: 6.5 / MAX: 7.79 MIN: 7.08 / MAX: 8.77 MIN: 7.32 / MAX: 9.13 MIN: 7.3 / MAX: 9.13 MIN: 7.51 / MAX: 9.44 MIN: 7.84 / MAX: 9.82 MIN: 7.96 / MAX: 9.95 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.02 0.04 0.06 0.08 0.1 0.049 0.049 0.072 0.076 0.081 0.085 0.089 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 58.7 171.3 190.7 Xeon Gold 6421N 57.3 170.9 190.9 EPYC 8324P - Power 22.9 126.4 144.1 EPYC 8324PN - Power 18.1 118.1 134.3 EPYC 8324P 19.1 107.4 119.6 EPYC 8324P - 155W 19.0 95.7 105.0 EPYC 8324PN 17.9 81.5 88.9 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 16.21 16.24 17.32 19.19 20.14 20.30 20.30 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.052 0.104 0.156 0.208 0.26 0.110 0.111 0.191 0.205 0.208 0.224 0.231 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 60.8 147.9 161.5 Xeon Gold 6421N - 6c RAM 55.5 146.5 160.3 EPYC 8324P - Power 18.0 106.4 123.6 EPYC 8324PN - Power 17.0 99.2 114.5 EPYC 8324P 18.1 96.8 115.8 EPYC 8324P - 155W 18.6 85.7 99.4 EPYC 8324PN 16.7 75.1 84.2 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 50 100 150 200 250
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Exhaustive EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 0.4862 0.9724 1.4586 1.9448 2.431 SE +/- 0.0005, N = 3 SE +/- 0.0011, N = 3 SE +/- 0.0005, N = 3 SE +/- 0.0004, N = 3 SE +/- 0.0003, N = 3 SE +/- 0.0023, N = 3 SE +/- 0.0008, N = 3 1.7295 1.9475 2.0103 2.0103 2.1194 2.1580 2.1608 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Exhaustive Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324P EPYC 8324PN 0.005 0.01 0.015 0.02 0.025 0.013 0.013 0.019 0.020 0.021 0.022 0.022 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 62.0 152.3 189.3 Xeon Gold 6421N - 6c RAM 54.1 152.2 188.6 EPYC 8324P - Power 19.1 112.2 143.5 EPYC 8324PN - Power 17.5 105.9 134.3 EPYC 8324P 17.9 97.1 125.1 EPYC 8324P - 155W 18.8 90.7 111.0 EPYC 8324PN 24.1 79.1 92.4 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 50 100 150 200 250
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: Blowfish Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 12K 24K 36K 48K 60K SE +/- 5.00, N = 3 SE +/- 5.33, N = 3 SE +/- 45.61, N = 3 SE +/- 13.33, N = 3 SE +/- 17.01, N = 3 SE +/- 6.51, N = 3 SE +/- 29.13, N = 3 44577 46376 47990 53523 55462 55520 55526 -lbz2 -lbz2 -lbz2 -lbz2 -lbz2 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: Blowfish Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 140 280 420 560 700 304.25 319.08 527.92 534.47 550.11 597.75 650.88 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 60.0 148.1 164.5 Xeon Gold 6421N 55.6 145.3 164.0 EPYC 8324P - Power 24.6 105.4 120.1 EPYC 8324P 19.6 105.3 120.5 EPYC 8324PN - Power 21.6 100.9 114.8 EPYC 8324P - 155W 19.4 89.5 100.9 EPYC 8324PN 16.8 73.7 83.9 OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor 50 100 150 200 250
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.
Result
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324P - Power EPYC 8324P Xeon Gold 6421N 90 180 270 360 450 SE +/- 3.65, N = 3 SE +/- 2.44, N = 3 SE +/- 1.76, N = 5 SE +/- 1.18, N = 7 SE +/- 0.63, N = 3 SE +/- 2.69, N = 3 SE +/- 5.79, N = 3 319.17 332.63 347.66 355.02 359.88 360.01 395.62 MIN: 21.73 / MAX: 5000 MIN: 18.26 / MAX: 5000 MIN: 20.6 / MAX: 5000 MIN: 23.22 / MAX: 4615.38 MIN: 23.68 / MAX: 4000 MIN: 22.81 / MAX: 5000 MIN: 21.74 / MAX: 6000 Queries Per Minute, Geo Mean Per Watt
OpenBenchmarking.org Queries Per Minute, Geo Mean Per Watt, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power 1.1284 2.2568 3.3852 4.5136 5.642 3.018 3.731 4.831 4.867 4.891 4.942 5.015 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 59.2 106.0 180.5 Xeon Gold 6421N - 6c RAM 55.6 105.8 186.9 EPYC 8324P - Power 20.1 73.6 133.7 EPYC 8324P 18.6 72.8 114.5 EPYC 8324P - 155W 19.8 71.4 102.4 EPYC 8324PN - Power 17.6 70.8 121.4 EPYC 8324PN 20.1 68.9 91.5 OpenBenchmarking.org Watts, Fewer Is Better ClickHouse 22.12.3.5 CPU Power Consumption Monitor 50 100 150 200 250
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 12.95 13.28 13.31 14.49 15.60 15.96 16.00 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.0374 0.0748 0.1122 0.1496 0.187 0.082 0.083 0.136 0.151 0.151 0.162 0.166 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 59.6 162.4 175.1 Xeon Gold 6421N - 6c RAM 56.1 160.6 175.0 EPYC 8324P - Power 18.1 117.3 133.4 EPYC 8324PN - Power 17.4 105.7 120.3 EPYC 8324P 33.3 103.1 114.0 EPYC 8324P - 155W 19.0 89.7 101.7 EPYC 8324PN 21.7 78.2 86.3 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 50 100 150 200 250
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.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N 80 160 240 320 400 SE +/- 0.39, N = 3 SE +/- 2.82, N = 3 SE +/- 2.25, N = 5 SE +/- 4.95, N = 3 SE +/- 2.87, N = 3 SE +/- 2.72, N = 7 SE +/- 3.12, N = 3 317.23 332.04 341.20 349.27 354.66 357.13 390.25 MIN: 21.7 / MAX: 5454.55 MIN: 18.33 / MAX: 4285.71 MIN: 20.94 / MAX: 4000 MIN: 22.61 / MAX: 4000 MIN: 23.68 / MAX: 3750 MIN: 23.5 / MAX: 5454.55 MIN: 21.65 / MAX: 5454.55
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N 80 160 240 320 400 SE +/- 2.14, N = 3 SE +/- 3.78, N = 3 SE +/- 3.40, N = 5 SE +/- 0.67, N = 3 SE +/- 2.16, N = 3 SE +/- 3.13, N = 7 SE +/- 2.58, N = 3 303.78 321.56 329.38 334.83 341.29 343.11 373.19 MIN: 21.61 / MAX: 6000 MIN: 18.26 / MAX: 5000 MIN: 20.59 / MAX: 4285.71 MIN: 22.37 / MAX: 4285.71 MIN: 23.46 / MAX: 4285.71 MIN: 23.22 / MAX: 5000 MIN: 21.45 / MAX: 4615.38
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Very Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 10 20 30 40 50 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 4 SE +/- 0.04, N = 4 SE +/- 0.04, N = 4 SE +/- 0.07, N = 4 SE +/- 0.02, N = 4 34.84 34.94 37.11 40.74 41.92 42.00 42.01 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Very Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.1233 0.2466 0.3699 0.4932 0.6165 0.271 0.275 0.472 0.474 0.521 0.529 0.548 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 61.8 129.1 149.2 Xeon Gold 6421N - 6c RAM 55.8 126.7 148.7 EPYC 8324P - Power 27.2 88.9 113.4 EPYC 8324P 18.4 88.5 111.6 EPYC 8324PN - Power 17.0 80.6 107.3 EPYC 8324P - 155W 20.1 77.0 95.9 EPYC 8324PN 25.8 67.7 81.3 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 40 80 120 160 200
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast EPYC 8324PN Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 10 20 30 40 50 SE +/- 0.02, N = 4 SE +/- 0.04, N = 4 SE +/- 0.04, N = 4 SE +/- 0.08, N = 4 SE +/- 0.02, N = 4 SE +/- 0.02, N = 4 SE +/- 0.02, N = 4 36.40 37.35 37.36 40.32 42.56 43.57 43.73 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.1181 0.2362 0.3543 0.4724 0.5905 0.264 0.268 0.455 0.497 0.499 0.515 0.525 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.0 141.4 168.5 Xeon Gold 6421N 56.6 139.3 170.1 EPYC 8324P - Power 18.8 96.1 127.0 EPYC 8324PN - Power 17.0 87.6 117.1 EPYC 8324P 18.8 85.2 114.7 EPYC 8324P - 155W 18.2 78.2 101.0 EPYC 8324PN 16.6 69.4 85.0 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 50 100 150 200 250
TiDB Community Server This is a PingCAP TiDB Community Server benchmark facilitated using the sysbench OLTP database benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Queries Per Second, More Is Better TiDB Community Server 7.3 Test: oltp_update_non_index - Threads: 64 EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 9K 18K 27K 36K 45K SE +/- 48.81, N = 3 SE +/- 83.64, N = 3 SE +/- 172.93, N = 3 SE +/- 104.02, N = 3 SE +/- 19.08, N = 3 SE +/- 76.77, N = 3 SE +/- 98.03, N = 3 33484 36088 36397 36662 36829 39692 40075 Queries Per Second Per Watt
OpenBenchmarking.org Queries Per Second Per Watt, More Is Better TiDB Community Server 7.3 Test: oltp_update_non_index - Threads: 64 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 110 220 330 440 550 354.40 356.46 447.36 453.83 467.32 472.02 490.43 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 59.0 112.4 149.0 Xeon Gold 6421N - 6c RAM 55.9 112.0 148.0 EPYC 8324P 19.0 81.4 112.2 EPYC 8324P - Power 18.7 80.8 111.6 EPYC 8324PN - Power 18.0 78.0 106.1 EPYC 8324P - 155W 18.8 77.2 107.2 EPYC 8324PN 18.5 68.3 81.5 OpenBenchmarking.org Watts, Fewer Is Better TiDB Community Server 7.3 CPU Power Consumption Monitor 40 80 120 160 200
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Medium Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324P - Power 10 20 30 40 50 SE +/- 0.09, N = 4 SE +/- 0.05, N = 4 SE +/- 0.08, N = 4 SE +/- 0.10, N = 4 SE +/- 0.07, N = 4 SE +/- 0.04, N = 4 SE +/- 0.05, N = 4 36.89 36.99 42.43 43.95 43.96 44.07 44.13 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Medium Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324PN 0.1575 0.315 0.4725 0.63 0.7875 0.344 0.346 0.602 0.612 0.630 0.645 0.700 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 59.0 107.1 122.9 Xeon Gold 6421N 55.3 107.0 122.5 EPYC 8324P 19.2 73.0 93.3 EPYC 8324P - 155W 19.0 72.1 92.2 EPYC 8324P - Power 18.7 70.0 92.1 EPYC 8324PN - Power 17.7 68.2 88.7 EPYC 8324PN 28.5 60.6 73.2 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 40 80 120 160 200
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.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 400M 800M 1200M 1600M 2000M SE +/- 1039764.93, N = 3 SE +/- 202758.75, N = 3 SE +/- 351188.46, N = 3 SE +/- 1733333.33, N = 3 SE +/- 1841496.98, N = 3 SE +/- 808977.41, N = 3 SE +/- 1017076.42, N = 3 1432866667 1575233333 1575300000 1603333333 1710466667 1713066667 1713766667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 4M 8M 12M 16M 20M 9744826.95 10070568.78 16129728.43 17155799.92 17329956.10 18295373.28 19130906.93 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 70.1 161.7 188.4 Xeon Gold 6421N - 6c RAM 57.8 156.4 183.3 EPYC 8324P - Power 19.7 106.2 124.4 EPYC 8324PN - Power 16.7 99.9 117.6 EPYC 8324P 19.9 98.7 113.4 EPYC 8324P - 155W 22.6 87.6 100.3 EPYC 8324PN 19.2 74.9 84.6 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 50 100 150 200 250
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 32 EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324P EPYC 8324P - Power 400M 800M 1200M 1600M 2000M SE +/- 3317797.13, N = 3 SE +/- 850490.05, N = 3 SE +/- 1433333.33, N = 3 SE +/- 2423037.58, N = 3 SE +/- 208166.60, N = 3 SE +/- 1649579.07, N = 3 SE +/- 2162046.36, N = 3 1446866667 1590300000 1593866667 1612933333 1726900000 1728433333 1729733333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid samples/s Per Watt
OpenBenchmarking.org samples/s Per Watt, More Is Better Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 32 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 4M 8M 12M 16M 20M 10118407.57 10154280.68 15497527.92 16825480.34 17231776.41 18252252.86 19663189.85 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 57.2 157.2 176.1 Xeon Gold 6421N 56.0 157.0 176.6 EPYC 8324P - Power 52.2 111.6 125.8 EPYC 8324PN - Power 17.3 102.6 118.8 EPYC 8324P 36.5 100.3 113.6 EPYC 8324P - 155W 21.1 88.4 101.4 EPYC 8324PN 17.1 73.6 85.1 OpenBenchmarking.org Watts, Fewer Is Better Liquid-DSP 1.6 CPU Power Consumption Monitor 50 100 150 200 250
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 9 18 27 36 45 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 4 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 SE +/- 0.02, N = 4 33.41 33.74 33.82 37.06 38.79 39.63 39.88 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.1085 0.217 0.3255 0.434 0.5425 0.236 0.243 0.415 0.433 0.445 0.449 0.482 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.5 143.0 166.0 Xeon Gold 6421N - 6c RAM 55.6 139.1 165.5 EPYC 8324P - Power 18.9 96.2 125.1 EPYC 8324P 18.4 89.5 113.0 EPYC 8324PN - Power 17.0 89.1 115.3 EPYC 8324P - 155W 19.7 82.5 100.1 EPYC 8324PN 17.4 69.3 84.1 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 50 100 150 200 250
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough EPYC 8324PN EPYC 8324P - 155W EPYC 8324P Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN - Power EPYC 8324P - Power 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 4 SE +/- 0.00, N = 4 17.31 19.34 20.40 20.54 20.54 20.56 20.61 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Thorough Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN 0.056 0.112 0.168 0.224 0.28 0.151 0.151 0.214 0.229 0.240 0.247 0.249 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 58.1 136.2 186.2 Xeon Gold 6421N 57.2 136.0 186.4 EPYC 8324P - Power 18.8 96.2 143.0 EPYC 8324PN - Power 17.5 89.8 134.1 EPYC 8324P 17.8 84.9 125.1 EPYC 8324P - 155W 17.6 78.2 110.2 EPYC 8324PN 17.1 69.6 93.7 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 50 100 150 200 250
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Super Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 12 24 36 48 60 SE +/- 0.04, N = 4 SE +/- 0.07, N = 4 SE +/- 0.02, N = 4 SE +/- 0.17, N = 4 SE +/- 0.03, N = 5 SE +/- 0.11, N = 5 SE +/- 0.05, N = 5 45.85 45.92 47.11 51.82 53.91 54.05 54.10 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Super Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.1692 0.3384 0.5076 0.6768 0.846 0.361 0.368 0.623 0.654 0.655 0.692 0.752 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 56.5 127.1 154.9 Xeon Gold 6421N - 6c RAM 58.3 124.7 153.8 EPYC 8324P - Power 19.4 86.8 116.1 EPYC 8324PN - Power 16.9 82.6 110.4 EPYC 8324P 19.2 82.4 112.9 EPYC 8324P - 155W 26.4 74.9 97.9 EPYC 8324PN 17.2 62.7 83.1 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 40 80 120 160 200
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.
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.
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 11 22 33 44 55 SE +/- 0.02, N = 4 SE +/- 0.06, N = 4 SE +/- 0.02, N = 4 SE +/- 0.05, N = 4 SE +/- 0.05, N = 4 SE +/- 0.03, N = 4 SE +/- 0.03, N = 4 40.68 42.48 42.53 44.70 46.20 47.08 47.35 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324PN EPYC 8324P - 155W 0.1343 0.2686 0.4029 0.5372 0.6715 0.312 0.317 0.508 0.534 0.556 0.588 0.597 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 71.9 136.3 160.7 Xeon Gold 6421N 60.8 134.3 161.9 EPYC 8324P - Power 29.7 93.1 119.4 EPYC 8324P 26.5 86.5 112.7 EPYC 8324PN - Power 17.1 84.6 110.3 EPYC 8324P - 155W 18.4 74.8 99.8 EPYC 8324PN 17.1 69.2 83.1 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 50 100 150 200 250
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324PN - Power 5 10 15 20 25 SE +/- 0.11, N = 5 SE +/- 0.17, N = 5 SE +/- 0.11, N = 5 SE +/- 0.09, N = 5 SE +/- 0.14, N = 5 SE +/- 0.19, N = 5 SE +/- 0.15, N = 5 17.13 18.60 19.39 19.57 19.71 19.79 19.90 MIN: 15.39 / MAX: 17.81 MIN: 16.6 / MAX: 19.37 MIN: 17.5 / MAX: 19.92 MIN: 17.74 / MAX: 20 MIN: 17.86 / MAX: 20.53 MIN: 17.75 / MAX: 20.75 MIN: 17.77 / MAX: 20.63 M samples/sec Per Watt
OpenBenchmarking.org M samples/sec Per Watt, More Is Better LuxCoreRender 2.6 Scene: Rainbow Colors and Prism - Acceleration: CPU Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.0655 0.131 0.1965 0.262 0.3275 0.135 0.142 0.232 0.233 0.256 0.266 0.291 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 81.8 146.0 190.0 Xeon Gold 6421N - 6c RAM 63.4 139.8 191.6 EPYC 8324P - Power 19.3 84.3 121.1 EPYC 8324P 19.1 83.1 120.4 EPYC 8324PN - Power 17.3 77.8 115.1 EPYC 8324P - 155W 23.5 70.0 99.5 EPYC 8324PN 17.1 58.9 82.9 OpenBenchmarking.org Watts, Fewer Is Better LuxCoreRender 2.6 CPU Power Consumption Monitor 50 100 150 200 250
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 30 60 90 120 150 SE +/- 0.11, N = 7 SE +/- 0.12, N = 7 SE +/- 0.09, N = 7 SE +/- 0.10, N = 7 SE +/- 0.12, N = 7 SE +/- 0.13, N = 7 SE +/- 0.12, N = 7 104.53 104.57 113.93 120.19 120.39 120.82 121.09 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 0.5072 1.0144 1.5216 2.0288 2.536 1.048 1.060 1.980 1.995 2.099 2.117 2.254 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 57.1 99.7 122.5 Xeon Gold 6421N 56.4 98.7 133.9 EPYC 8324P 18.9 60.8 97.9 EPYC 8324P - Power 17.8 60.6 92.8 EPYC 8324P - 155W 17.5 57.3 95.5 EPYC 8324PN - Power 18.4 57.2 94.5 EPYC 8324PN 19.9 50.6 78.2 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 40 80 120 160 200
SVT-AV1 Result
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K EPYC 8324PN Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W Xeon Gold 6421N EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power 20 40 60 80 100 SE +/- 0.08, N = 4 SE +/- 0.28, N = 5 SE +/- 0.14, N = 4 SE +/- 0.29, N = 5 SE +/- 0.14, N = 5 SE +/- 0.09, N = 5 SE +/- 0.23, N = 5 69.46 74.50 74.62 74.69 78.70 79.38 80.05 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 0.2723 0.5446 0.8169 1.0892 1.3615 0.648 0.661 1.113 1.151 1.179 1.192 1.210 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.1 115.2 167.0 Xeon Gold 6421N - 6c RAM 56.8 112.8 167.1 EPYC 8324P - Power 15.2 71.9 124.3 EPYC 8324P 17.4 68.4 112.3 EPYC 8324PN - Power 17.2 66.6 113.4 EPYC 8324P - 155W 19.6 63.3 98.3 EPYC 8324PN 17.9 57.4 84.9 OpenBenchmarking.org Watts, Fewer Is Better SVT-AV1 1.7 CPU Power Consumption Monitor 50 100 150 200 250
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power 30 60 90 120 150 SE +/- 0.12, N = 7 SE +/- 0.20, N = 7 SE +/- 0.06, N = 8 SE +/- 0.08, N = 8 SE +/- 0.14, N = 8 SE +/- 0.10, N = 8 SE +/- 0.08, N = 8 119.12 119.20 127.09 134.92 136.34 136.57 137.07 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 0.605 1.21 1.815 2.42 3.025 1.196 1.220 2.235 2.324 2.345 2.394 2.689 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.1 99.7 135.8 Xeon Gold 6421N - 6c RAM 57.7 97.6 136.6 EPYC 8324P - Power 26.3 61.0 93.3 EPYC 8324P 19.2 58.8 92.9 EPYC 8324P - 155W 20.1 57.5 89.3 EPYC 8324PN - Power 17.6 57.3 89.2 EPYC 8324PN 16.3 47.3 72.1 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 40 80 120 160 200
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: HMAC-SHA512 EPYC 8324PN EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N - 6c RAM Xeon Gold 6421N 40M 80M 120M 160M 200M SE +/- 269095.40, N = 3 SE +/- 642230.83, N = 3 SE +/- 203030.65, N = 3 SE +/- 414666.26, N = 3 SE +/- 29597.30, N = 3 SE +/- 421171.51, N = 3 SE +/- 320167.84, N = 3 163350000 170626333 171710333 171762667 171768000 174776333 186859333 -lbz2 -lbz2 -lbz2 -lbz2 -lbz2 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: HMAC-SHA512 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 500K 1000K 1500K 2000K 2500K 1281524.49 1338344.69 2075055.26 2078692.75 2134574.10 2195515.23 2260685.57 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 58.2 139.6 163.1 Xeon Gold 6421N - 6c RAM 62.3 136.4 162.4 EPYC 8324P - Power 19.3 82.6 105.0 EPYC 8324P 18.4 82.2 105.1 EPYC 8324P - 155W 18.7 80.4 104.2 EPYC 8324PN - Power 17.8 78.2 100.0 EPYC 8324PN 17.5 72.3 85.5 OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor 50 100 150 200 250
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.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 50K 100K 150K 200K 250K SE +/- 114.10, N = 3 SE +/- 119.68, N = 3 SE +/- 222.34, N = 3 SE +/- 162.02, N = 3 SE +/- 132.07, N = 3 SE +/- 340.61, N = 3 SE +/- 265.09, N = 3 211839 218723 227218 228756 239650 240409 241268 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium EPYC 8324PN Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 40 80 120 160 200 SE +/- 0.06, N = 6 SE +/- 0.02, N = 6 SE +/- 0.02, N = 6 SE +/- 0.05, N = 6 SE +/- 0.03, N = 6 SE +/- 0.05, N = 6 SE +/- 0.07, N = 6 141.98 150.78 150.81 154.14 160.75 161.22 161.44 1. (CXX) g++ options: -O3 -flto -pthread MT/s Per Watt
OpenBenchmarking.org MT/s Per Watt, More Is Better ASTC Encoder 4.0 Preset: Medium Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324PN EPYC 8324P - 155W 0.6298 1.2596 1.8894 2.5192 3.149 1.449 1.464 2.590 2.732 2.781 2.786 2.799 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.2 104.1 152.6 Xeon Gold 6421N - 6c RAM 58.0 103.0 151.9 EPYC 8324P - Power 17.7 62.2 106.1 EPYC 8324P 18.4 58.8 98.4 EPYC 8324PN - Power 17.4 58.1 98.7 EPYC 8324P - 155W 17.4 55.1 90.8 EPYC 8324PN 17.1 51.0 81.6 OpenBenchmarking.org Watts, Fewer Is Better ASTC Encoder 4.0 CPU Power Consumption Monitor 40 80 120 160 200
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.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - 155W EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 20000M 40000M 60000M 80000M 100000M SE +/- 219080756.78, N = 3 SE +/- 28304630.52, N = 3 SE +/- 23664285.60, N = 3 SE +/- 7496950.89, N = 3 SE +/- 5790417.71, N = 3 SE +/- 1978138.59, N = 3 SE +/- 6320340.94, N = 3 99907867353 106988442053 107133667010 111478070750 113098144337 113108043093 113119629900 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 200M 400M 600M 800M 1000M 668230615.40 679280895.67 903365427.29 915346369.78 963770455.72 1040034038.92 1139876908.44 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 63.7 160.1 176.6 Xeon Gold 6421N 59.1 157.7 171.7 EPYC 8324P - Power 34.2 125.2 139.7 EPYC 8324P 39.0 123.6 130.9 EPYC 8324PN - Power 17.1 117.4 131.4 EPYC 8324P - 155W 18.9 107.2 112.8 EPYC 8324PN 34.1 87.6 91.7 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100 EPYC 8324PN Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324P - Power EPYC 8324P Xeon Gold 6421N 20M 40M 60M 80M 100M SE +/- 447525.63, N = 3 SE +/- 326677.77, N = 3 SE +/- 719780.74, N = 3 SE +/- 386916.43, N = 3 SE +/- 748207.16, N = 3 SE +/- 281149.50, N = 3 SE +/- 445547.83, N = 3 89411612 92766907 94808492 97338332 97619880 98984707 100840933
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100 EPYC 8324PN Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power EPYC 8324P - Power Xeon Gold 6421N 20M 40M 60M 80M 100M SE +/- 577853.90, N = 3 SE +/- 283803.69, N = 3 SE +/- 539297.25, N = 3 SE +/- 94386.99, N = 3 SE +/- 469635.66, N = 3 SE +/- 396005.49, N = 3 SE +/- 136377.33, N = 3 89550933 91362622 93050746 93807432 94636861 96383625 100615762
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324PN EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P 14 28 42 56 70 SE +/- 0.04, N = 5 SE +/- 0.14, N = 5 SE +/- 0.08, N = 5 SE +/- 0.07, N = 5 SE +/- 0.06, N = 5 SE +/- 0.19, N = 5 SE +/- 0.08, N = 5 55.45 55.60 56.06 61.09 61.93 62.09 62.18 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 0.2126 0.4252 0.6378 0.8504 1.063 0.469 0.493 0.805 0.815 0.836 0.864 0.945 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 56.7 118.2 143.6 Xeon Gold 6421N 56.4 112.8 143.5 EPYC 8324P 18.4 77.2 107.3 EPYC 8324P - Power 19.3 76.0 106.4 EPYC 8324PN - Power 17.9 74.3 102.4 EPYC 8324P - 155W 19.7 70.7 93.1 EPYC 8324PN 17.1 59.3 78.4 OpenBenchmarking.org Watts, Fewer Is Better Kvazaar 2.2 CPU Power Consumption Monitor 40 80 120 160 200
uvg266 uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324PN EPYC 8324P EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324PN - Power 30 60 90 120 150 SE +/- 0.18, N = 8 SE +/- 0.11, N = 8 SE +/- 0.11, N = 8 SE +/- 0.11, N = 8 SE +/- 0.09, N = 8 SE +/- 0.17, N = 8 SE +/- 0.11, N = 8 134.08 134.18 141.61 148.70 148.95 149.14 149.56 Frames Per Second Per Watt
OpenBenchmarking.org Frames Per Second Per Watt, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324P - 155W EPYC 8324PN - Power EPYC 8324PN 0.6761 1.3522 2.0283 2.7044 3.3805 1.424 1.435 2.605 2.626 2.677 2.796 3.005 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 58.2 94.2 124.1 Xeon Gold 6421N 59.6 93.5 123.1 EPYC 8324P - Power 17.3 57.3 82.8 EPYC 8324P 18.1 56.6 83.9 EPYC 8324P - 155W 19.0 55.6 83.0 EPYC 8324PN - Power 17.4 53.5 79.7 EPYC 8324PN 17.3 47.1 69.7 OpenBenchmarking.org Watts, Fewer Is Better uvg266 0.4.1 CPU Power Consumption Monitor 40 80 120 160 200
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100 EPYC 8324PN Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power Xeon Gold 6421N 20M 40M 60M 80M 100M SE +/- 34230.69, N = 3 SE +/- 367599.31, N = 3 SE +/- 708277.70, N = 3 SE +/- 553005.37, N = 3 SE +/- 271846.31, N = 3 SE +/- 811868.15, N = 3 SE +/- 1085408.93, N = 3 96286895 101011034 101322673 104603173 104847158 105469536 106918573
John The Ripper This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Real C/S, More Is Better John The Ripper 2023.03.14 Test: WPA PSK EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power 40K 80K 120K 160K 200K SE +/- 250.16, N = 3 SE +/- 571.00, N = 3 SE +/- 21.33, N = 3 SE +/- 99.38, N = 3 SE +/- 113.67, N = 3 SE +/- 0.33, N = 3 SE +/- 78.81, N = 3 183737 185188 192699 202775 202843 203025 203093 -lbz2 -lbz2 -lbz2 -lbz2 -lbz2 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt Real C/S Per Watt
OpenBenchmarking.org Real C/S Per Watt, More Is Better John The Ripper 2023.03.14 Test: WPA PSK Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 500 1000 1500 2000 2500 1333.94 1379.60 1797.39 1799.81 1840.93 1945.44 2203.70 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 56.9 139.7 156.9 Xeon Gold 6421N - 6c RAM 62.2 138.8 155.8 EPYC 8324P - Power 20.2 112.8 132.6 EPYC 8324P 18.4 112.8 133.0 EPYC 8324PN - Power 46.3 110.3 124.8 EPYC 8324P - 155W 19.7 104.3 120.3 EPYC 8324PN 17.8 83.4 96.1 OpenBenchmarking.org Watts, Fewer Is Better John The Ripper 2023.03.14 CPU Power Consumption Monitor 40 80 120 160 200
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.
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 EPYC 8324PN Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P - 155W EPYC 8324P EPYC 8324PN - Power 30000M 60000M 90000M 120000M 150000M SE +/- 348076081.82, N = 3 SE +/- 44088659.43, N = 3 SE +/- 3290643.90, N = 3 SE +/- 19688066.61, N = 3 SE +/- 3097156.83, N = 3 SE +/- 1094343.67, N = 3 SE +/- 3489061.39, N = 3 146313390400 149077028223 149212297717 159433379203 159437074800 159441828337 159458412410 -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto -lssl -lcrypto 1. (CC) gcc options: -pthread -m64 -O3 -ldl byte/s Per Watt
OpenBenchmarking.org byte/s Per Watt, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 Xeon Gold 6421N - 6c RAM Xeon Gold 6421N EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 400M 800M 1200M 1600M 2000M 952146897.19 964354097.71 1311612149.78 1326792043.15 1383242881.11 1437696738.22 1668421471.53 CPU Power Consumption
Min Avg Max Xeon Gold 6421N - 6c RAM 68.9 156.6 185.9 Xeon Gold 6421N 55.6 154.7 184.8 EPYC 8324P - Power 19.4 121.6 138.3 EPYC 8324P 19.3 120.2 129.6 EPYC 8324PN - Power 17.6 115.3 129.8 EPYC 8324P - 155W 19.3 110.9 116.7 EPYC 8324PN 22.7 87.7 93.6 OpenBenchmarking.org Watts, Fewer Is Better OpenSSL 3.1 CPU Power Consumption Monitor 50 100 150 200 250
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power EPYC 8324P - 155W EPYC 8324PN 40 80 120 160 200 Min: 46.98 / Avg: 139.29 / Max: 192.53 Min: 49.5 / Avg: 137 / Max: 192.41 Min: 15.15 / Avg: 94.3 / Max: 168.07 Min: 15 / Avg: 89.8 / Max: 136.05 Min: 13.98 / Avg: 88.14 / Max: 134.59 Min: 15.85 / Avg: 82.68 / Max: 120.33 Min: 14.57 / Avg: 72.32 / Max: 103.57
YugabyteDB Min Avg Max Xeon Gold 6421N 57.3 114.4 162.1 Xeon Gold 6421N - 6c RAM 60.9 113.7 156.1 EPYC 8324P - 155W 20.8 75.7 105.1 EPYC 8324P - Power 19.3 69.4 107.5 EPYC 8324P 17.4 67.7 107.8 EPYC 8324PN - Power 22.2 64.7 103.5 EPYC 8324PN 18.4 64.6 82.9 OpenBenchmarking.org Watts, Fewer Is Better YugabyteDB 2.19 CPU Power Consumption Monitor 50 100 150 200 250
OpenBenchmarking.org Read Ops/sec Per Watt, More Is Better YugabyteDB 2.19 Workload: CassandraBatchKeyValue, Batch 25 - Num Threads Write: 32 - Num Threads Read: 32 Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324PN EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power 400 800 1200 1600 2000 853.05 902.10 999.55 1092.66 1583.73 1607.06 1704.89
Result
OpenBenchmarking.org Read Ops/sec, More Is Better YugabyteDB 2.19 Workload: CassandraBatchKeyValue, Batch 25 - Num Threads Write: 32 - Num Threads Read: 32 EPYC 8324PN EPYC 8324P - 155W Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P EPYC 8324P - Power EPYC 8324PN - Power 20K 40K 60K 80K 100K SE +/- 5137.67, N = 12 SE +/- 6726.32, N = 15 SE +/- 4101.45, N = 15 SE +/- 5301.00, N = 15 SE +/- 325.58, N = 3 SE +/- 1324.92, N = 4 SE +/- 1184.23, N = 3 70619.32 75678.99 96623.75 102607.75 108862.05 109894.39 110313.78 Read Ops/sec Per Watt
OpenBenchmarking.org Read Ops/sec Per Watt, More Is Better YugabyteDB 2.19 Workload: CassandraBatchKeyValue, Batch 25 - Num Threads Write: 32 - Num Threads Read: 32 Xeon Gold 6421N Xeon Gold 6421N - 6c RAM EPYC 8324P - 155W EPYC 8324PN EPYC 8324P - Power EPYC 8324P EPYC 8324PN - Power 400 800 1200 1600 2000 853.05 902.10 999.55 1092.66 1583.73 1607.06 1704.89 CPU Power Consumption
Min Avg Max Xeon Gold 6421N 57.3 114.4 162.1 Xeon Gold 6421N - 6c RAM 60.9 113.7 156.1 EPYC 8324P - 155W 20.8 75.7 105.1 EPYC 8324P - Power 19.3 69.4 107.5 EPYC 8324P 17.4 67.7 107.8 EPYC 8324PN - Power 22.2 64.7 103.5 EPYC 8324PN 18.4 64.6 82.9 OpenBenchmarking.org Watts, Fewer Is Better YugabyteDB 2.19 CPU Power Consumption Monitor 50 100 150 200 250
EPYC 8324P Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 19 September 2023 14:18 by user root.
EPYC 8324P - 155W Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 21 September 2023 20:07 by user root.
EPYC 8324P - Power Processor: AMD EPYC 8324P 32-Core @ 2.65GHz (32 Cores / 64 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32 GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: llvmpipe, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7, OpenGL: 4.5 Mesa 23.1.7-1ubuntu1 (LLVM 15.0.7 256 bits), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 640x480
Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 21 September 2023 00:52 by user root.
EPYC 8324PN Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 22 September 2023 22:47 by user root.
EPYC 8324PN - Power Processor: AMD EPYC 8534PN 32-Core @ 2.05GHz (32 Cores / 64 Threads), Motherboard: AMD Cinnabar (RCB1009C BIOS), Chipset: AMD Device 14a4, Memory: 6 x 32 GB DDR5-4800MT/s Samsung M321R4GA0BB0-CQKMG, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.7, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 640x480
Kernel Notes: Transparent Huge Pages: madviseCompiler 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-FTCNCZ/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-FTCNCZ/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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xaa00212Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 23 September 2023 16:16 by user root.
Xeon Gold 6421N Processor: Intel Xeon Gold 6421N @ 3.60GHz (32 Cores / 64 Threads), Motherboard: ASRockRack SPC741D8-2L2T/BCM (1.26 BIOS), Chipset: Intel Device 1bce, Memory: 8 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: 2 x Intel I210 + 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (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: 640x480
Kernel Notes: Transparent Huge Pages: madviseCompiler 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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 1 October 2023 00:30 by user root.
Xeon Gold 6421N - 6c RAM Processor: Intel Xeon Gold 6421N @ 3.60GHz (32 Cores / 64 Threads), Motherboard: ASRockRack SPC741D8-2L2T/BCM (1.26 BIOS), Chipset: Intel Device 1bce, Memory: 6 x 32 GB DDR5-4400MT/s Micron MTC20F2085S1RC48BA1, Disk: 800GB Micron_7450_MTFDKBA800TFS + 4 x 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: 2 x Intel I210 + 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA
OS: Ubuntu 23.10, Kernel: 6.5.0-5-generic (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: 640x480
Kernel Notes: Transparent Huge Pages: madviseCompiler 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 -vDisk Notes: NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Notes: Scaling Governor: intel_pstate performance (EPP: performance) - CPU Microcode: 0x2b0004b1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu1)Python Notes: Python 3.11.5Security 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 1 October 2023 22:40 by user root.