Benchmarks by Michael Larabel for a future article.
Auto IBRS Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.2.0-rc3-phx-tip-x86-cpu (x86_64), Desktop: GNOME Shell 43.1, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2)Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Linux 6.2 Git OS: Ubuntu 22.10, Kernel: 6.2.0-060200rc4daily20230121-generic (x86_64), Desktop: GNOME Shell 43.1, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2)Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
AMD EPYC Genoa Auto IBRS Benchmark OpenBenchmarking.org Phoronix Test Suite 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads) AMD Titanite_4G (RTI1002E BIOS) AMD Device 14a4 1520GB 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007 ASPEED VGA HDMI Broadcom NetXtreme BCM5720 PCIe Ubuntu 22.10 6.2.0-rc3-phx-tip-x86-cpu (x86_64) 6.2.0-060200rc4daily20230121-generic (x86_64) GNOME Shell 43.1 X Server 1.21.1.4 1.3.224 GCC 12.2.0 ext4 1920x1080 Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernels Desktop Display Server Vulkan Compiler File-System Screen Resolution AMD EPYC Genoa Auto IBRS Benchmark Performance System Logs - Transparent Huge Pages: madvise - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 - none / errors=remount-ro,relatime,rw / Block Size: 4096 - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110d - OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2) - Python 3.10.7 - Auto IBRS: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Linux 6.2 Git: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Auto IBRS vs. Linux 6.2 Git Comparison Phoronix Test Suite Baseline +2.3% +2.3% +4.6% +4.6% +6.9% +6.9% 100 - 1000 - Read Only - Average Latency 9.1% NUMA 9.1% 100 - 1000 - Read Only 8.9% Compile 8.8% 1000 - 1000 - Read Only - Average Latency 8.1% Throughput 7.7% Initial Create 7.6% 1000 - 1000 - Read Only 7.4% Latency Under Load 7.2% SENDFILE 5.7% 100 - 800 - Read Only 5% 100 - 800 - Read Only - Average Latency 4.9% Rand Write - IO_uring - No - Yes - 4KB 4.5% 4.4% Scala Dotty 4.3% Latency Ping Pong 3.8% Rand Fill 3.7% Sched Pipe 3.6% 500 3.1% D.T.P 2.8% Writes 2.4% Mutex 2.4% Malloc 2.4% 40000000 - 100 - C.P.B 2.2% MoVR - 1024 2.1% Forking 2.1% PostgreSQL Stress-NG PostgreSQL Compile Bench PostgreSQL Sockperf Compile Bench PostgreSQL Sockperf Stress-NG PostgreSQL PostgreSQL Flexible IO Tester WireGuard + Linux Networking Stack Stress Test Renaissance Sockperf RocksDB perf-bench nginx PostMark Apache Cassandra Stress-NG Stress-NG Apache Spark CockroachDB Stress-NG Auto IBRS Linux 6.2 Git
AMD EPYC Genoa Auto IBRS Benchmark minibude: OpenMP - BM1 minibude: OpenMP - BM2 stress-ng: NUMA stress-ng: MEMFD stress-ng: Mutex stress-ng: Crypto stress-ng: Malloc stress-ng: Forking stress-ng: SENDFILE nekrs: TurboPipe Periodic perf-bench: Memcpy 1MB perf-bench: Memset 1MB minibude: OpenMP - BM1 minibude: OpenMP - BM2 fio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directory compilebench: Compile compilebench: Initial Create compilebench: Read Compiled Tree sockperf: Throughput cassandra: Writes rocksdb: Rand Fill rocksdb: Rand Read rocksdb: Update Rand cockroach: MoVR - 1024 cockroach: KV, 60% Reads - 512 cockroach: KV, 95% Reads - 512 cockroach: KV, 10% Reads - 1024 cockroach: KV, 60% Reads - 1024 perf-bench: Epoll Wait perf-bench: Futex Hash perf-bench: Sched Pipe perf-bench: Futex Lock-Pi nginx: 500 postmark: Disk Transaction Performance pgbench: 100 - 800 - Read Only pgbench: 100 - 1000 - Read Only pgbench: 1000 - 1000 - Read Only ctx-clock: Context Switch Time renaissance: Scala Dotty renaissance: Rand Forest renaissance: ALS Movie Lens renaissance: Apache Spark Bayes renaissance: Apache Spark PageRank renaissance: Finagle HTTP Requests renaissance: In-Memory Database Shootout renaissance: Akka Unbalanced Cobwebbed Tree renaissance: Genetic Algorithm Using Jenetics + Futures pgbench: 100 - 800 - Read Only - Average Latency pgbench: 100 - 1000 - Read Only - Average Latency pgbench: 1000 - 1000 - Read Only - Average Latency dacapobench: Jython wireguard: build-godot: Time To Compile build-linux-kernel: defconfig build-linux-kernel: allmodconfig build-llvm: Ninja build-nodejs: Time To Compile spark: 1000000 - 100 - SHA-512 Benchmark Time spark: 1000000 - 100 - Calculate Pi Benchmark spark: 40000000 - 100 - SHA-512 Benchmark Time spark: 40000000 - 100 - Calculate Pi Benchmark gpaw: Carbon Nanotube sockperf: Latency Ping Pong sockperf: Latency Under Load Auto IBRS Linux 6.2 Git 291.986 344.101 1033.80 2497.09 100349869.80 314759.54 666371878.04 4528.01 3312202.36 819373250000 42.561675 86.629582 7299.654 8602.519 417308 2518.56 674.77 3747.27 784554 261233 476505 899151523 468146 957.6 51541.0 63513.2 36055.8 52215.9 1085 2669764 381870 34 208165.81 6944 3538386 3498328 904587 144 674.8 636.5 20215.5 502.7 3176.6 12279.4 4743.9 29811.6 1716.7 0.226 0.286 1.106 3401 344.286 33.796 24.060 130.499 75.320 101.454 3.81 9.27 34.36 9.24 23.072 2.533 11.837 291.659 343.322 947.72 2487.90 98006797.30 313742.63 650976417.25 4433.77 3132228.87 820274333333 42.008348 87.378373 7291.489 8583.044 399250 2315.72 627.30 3695.65 728155 255089 459682 901191127 464816 937.6 51039.4 62732.2 35798.0 51688.7 1079 2675086 368457 34 201981.65 6756 3370717 3212970 842561 144 703.9 643.1 20043.9 506.0 3155.8 12401.5 4812.7 29879.9 1726.2 0.237 0.312 1.196 3352 359.458 34.237 24.285 131.919 75.424 101.966 3.82 9.38 34.65 9.44 23.171 2.630 12.694 OpenBenchmarking.org
miniBUDE MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 Auto IBRS Linux 6.2 Git 60 120 180 240 300 SE +/- 0.32, N = 3 SE +/- 0.48, N = 3 291.99 291.66 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.org Billion Interactions/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Auto IBRS Linux 6.2 Git 70 140 210 280 350 SE +/- 1.23, N = 3 SE +/- 3.26, N = 3 344.10 343.32 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MEMFD Auto IBRS Linux 6.2 Git 500 1000 1500 2000 2500 SE +/- 4.61, N = 3 SE +/- 1.36, N = 3 2497.09 2487.90 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Mutex Auto IBRS Linux 6.2 Git 20M 40M 60M 80M 100M SE +/- 1219212.33, N = 4 SE +/- 238740.03, N = 3 100349869.80 98006797.30 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Crypto Auto IBRS Linux 6.2 Git 70K 140K 210K 280K 350K SE +/- 319.91, N = 3 SE +/- 1383.90, N = 3 314759.54 313742.63 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Malloc Auto IBRS Linux 6.2 Git 140M 280M 420M 560M 700M SE +/- 5142965.01, N = 13 SE +/- 952566.38, N = 3 666371878.04 650976417.25 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Forking Auto IBRS Linux 6.2 Git 1000 2000 3000 4000 5000 SE +/- 2.29, N = 3 SE +/- 12.55, N = 3 4528.01 4433.77 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: SENDFILE Auto IBRS Linux 6.2 Git 700K 1400K 2100K 2800K 3500K SE +/- 11243.33, N = 3 SE +/- 29584.58, N = 3 3312202.36 3132228.87 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread
nekRS nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FLOP/s, More Is Better nekRS 22.0 Input: TurboPipe Periodic Auto IBRS Linux 6.2 Git 200000M 400000M 600000M 800000M 1000000M SE +/- 9162146595.05, N = 4 SE +/- 979478829.67, N = 3 819373250000 820274333333 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -lmpi
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memcpy 1MB Auto IBRS Linux 6.2 Git 10 20 30 40 50 SE +/- 0.07, N = 3 SE +/- 0.37, N = 3 42.56 42.01 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memset 1MB Auto IBRS Linux 6.2 Git 20 40 60 80 100 SE +/- 1.13, N = 3 SE +/- 0.42, N = 3 86.63 87.38 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
miniBUDE MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM1 Auto IBRS Linux 6.2 Git 1600 3200 4800 6400 8000 SE +/- 7.89, N = 3 SE +/- 11.94, N = 3 7299.65 7291.49 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.org GFInst/s, More Is Better miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Auto IBRS Linux 6.2 Git 2K 4K 6K 8K 10K SE +/- 30.74, N = 3 SE +/- 81.60, N = 3 8602.52 8583.04 1. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.29 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Auto IBRS Linux 6.2 Git 90K 180K 270K 360K 450K SE +/- 3752.58, N = 13 SE +/- 3473.11, N = 8 417308 399250 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile Auto IBRS Linux 6.2 Git 500 1000 1500 2000 2500 SE +/- 11.11, N = 3 SE +/- 4.51, N = 3 2518.56 2315.72
Sockperf This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.7 Test: Throughput Auto IBRS Linux 6.2 Git 200K 400K 600K 800K 1000K SE +/- 3740.42, N = 5 SE +/- 1132.22, N = 5 784554 728155 1. (CXX) g++ options: --param -O3 -rdynamic
RocksDB This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Fill Auto IBRS Linux 6.2 Git 100K 200K 300K 400K 500K SE +/- 6850.01, N = 3 SE +/- 2437.22, N = 3 476505 459682 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Read Auto IBRS Linux 6.2 Git 200M 400M 600M 800M 1000M SE +/- 557635.96, N = 3 SE +/- 471910.13, N = 3 899151523 901191127 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Update Random Auto IBRS Linux 6.2 Git 100K 200K 300K 400K 500K SE +/- 4332.00, N = 3 SE +/- 1841.36, N = 3 468146 464816 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
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.
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: MoVR - Concurrency: 1024 Auto IBRS Linux 6.2 Git 200 400 600 800 1000 SE +/- 0.56, N = 3 SE +/- 3.01, N = 3 957.6 937.6
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 512 Auto IBRS Linux 6.2 Git 11K 22K 33K 44K 55K SE +/- 118.90, N = 3 SE +/- 400.51, N = 15 51541.0 51039.4
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 512 Auto IBRS Linux 6.2 Git 14K 28K 42K 56K 70K SE +/- 731.10, N = 15 SE +/- 1022.03, N = 15 63513.2 62732.2
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 1024 Auto IBRS Linux 6.2 Git 8K 16K 24K 32K 40K SE +/- 382.05, N = 3 SE +/- 171.20, N = 3 36055.8 35798.0
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 1024 Auto IBRS Linux 6.2 Git 11K 22K 33K 44K 55K SE +/- 186.82, N = 3 SE +/- 146.61, N = 3 52215.9 51688.7
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Epoll Wait Auto IBRS Linux 6.2 Git 200 400 600 800 1000 SE +/- 8.19, N = 3 SE +/- 2.60, N = 3 1085 1079 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Hash Auto IBRS Linux 6.2 Git 600K 1200K 1800K 2400K 3000K SE +/- 400.67, N = 3 SE +/- 319.83, N = 3 2669764 2675086 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Sched Pipe Auto IBRS Linux 6.2 Git 80K 160K 240K 320K 400K SE +/- 2073.46, N = 3 SE +/- 1158.08, N = 3 381870 368457 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Lock-Pi Auto IBRS Linux 6.2 Git 8 16 24 32 40 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 34 34 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lz -llzma -lzstd -lnuma
nginx This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 Auto IBRS Linux 6.2 Git 40K 80K 120K 160K 200K SE +/- 111.07, N = 3 SE +/- 162.29, N = 3 208165.81 201981.65 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
PostMark This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance Auto IBRS Linux 6.2 Git 1500 3000 4500 6000 7500 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6944 6756 1. (CC) gcc options: -O3
PostgreSQL This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 41831.22, N = 3 SE +/- 34401.10, N = 3 3538386 3370717 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only Auto IBRS Linux 6.2 Git 700K 1400K 2100K 2800K 3500K SE +/- 9913.35, N = 3 SE +/- 46738.46, N = 12 3498328 3212970 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Only Auto IBRS Linux 6.2 Git 200K 400K 600K 800K 1000K SE +/- 3859.82, N = 3 SE +/- 23228.57, N = 12 904587 842561 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Random Forest Auto IBRS Linux 6.2 Git 140 280 420 560 700 SE +/- 6.07, N = 3 SE +/- 3.44, N = 3 636.5 643.1 MIN: 555.63 / MAX: 776.2 MIN: 567.13 / MAX: 785.73
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: ALS Movie Lens Auto IBRS Linux 6.2 Git 4K 8K 12K 16K 20K SE +/- 154.23, N = 3 SE +/- 306.67, N = 9 20215.5 20043.9 MIN: 19787 / MAX: 25483.46 MIN: 18228.15 / MAX: 26622.45
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Apache Spark Bayes Auto IBRS Linux 6.2 Git 110 220 330 440 550 SE +/- 3.95, N = 9 SE +/- 6.19, N = 3 502.7 506.0 MIN: 278.9 / MAX: 1571.68 MIN: 287.53 / MAX: 635.37
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Apache Spark PageRank Auto IBRS Linux 6.2 Git 700 1400 2100 2800 3500 SE +/- 33.91, N = 4 SE +/- 33.26, N = 15 3176.6 3155.8 MIN: 2696.76 / MAX: 3320.33 MIN: 2497.83 / MAX: 3643.1
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Finagle HTTP Requests Auto IBRS Linux 6.2 Git 3K 6K 9K 12K 15K SE +/- 134.77, N = 4 SE +/- 95.23, N = 3 12279.4 12401.5 MIN: 11111.23 / MAX: 12665.49 MIN: 11402.44 / MAX: 12591.71
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: In-Memory Database Shootout Auto IBRS Linux 6.2 Git 1000 2000 3000 4000 5000 SE +/- 51.76, N = 5 SE +/- 37.60, N = 10 4743.9 4812.7 MIN: 4294.27 / MAX: 6462.38 MIN: 4373 / MAX: 6522.68
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Akka Unbalanced Cobwebbed Tree Auto IBRS Linux 6.2 Git 6K 12K 18K 24K 30K SE +/- 365.89, N = 3 SE +/- 168.34, N = 3 29811.6 29879.9 MIN: 18792.65 / MAX: 30543.28 MIN: 17456.61 / MAX: 30119.88
OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Genetic Algorithm Using Jenetics + Futures Auto IBRS Linux 6.2 Git 400 800 1200 1600 2000 SE +/- 19.86, N = 4 SE +/- 11.38, N = 14 1716.7 1726.2 MIN: 1578.68 / MAX: 1867.26 MIN: 1497.5 / MAX: 1831.67
PostgreSQL This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.0533 0.1066 0.1599 0.2132 0.2665 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 0.226 0.237 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.0702 0.1404 0.2106 0.2808 0.351 SE +/- 0.001, N = 3 SE +/- 0.005, N = 12 0.286 0.312 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.2691 0.5382 0.8073 1.0764 1.3455 SE +/- 0.005, N = 3 SE +/- 0.031, N = 12 1.106 1.196 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
WireGuard + Linux Networking Stack Stress Test This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test Auto IBRS Linux 6.2 Git 80 160 240 320 400 SE +/- 3.62, N = 3 SE +/- 0.88, N = 3 344.29 359.46
Apache Spark This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time Auto IBRS Linux 6.2 Git 0.8595 1.719 2.5785 3.438 4.2975 SE +/- 0.03, N = 8 SE +/- 0.04, N = 5 3.81 3.82
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Auto IBRS Linux 6.2 Git 3 6 9 12 15 SE +/- 0.08, N = 8 SE +/- 0.12, N = 5 9.27 9.38
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - SHA-512 Benchmark Time Auto IBRS Linux 6.2 Git 8 16 24 32 40 SE +/- 0.36, N = 12 SE +/- 0.23, N = 3 34.36 34.65
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Calculate Pi Benchmark Auto IBRS Linux 6.2 Git 3 6 9 12 15 SE +/- 0.05, N = 12 SE +/- 0.02, N = 3 9.24 9.44
GPAW GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 22.1 Input: Carbon Nanotube Auto IBRS Linux 6.2 Git 6 12 18 24 30 SE +/- 0.07, N = 3 SE +/- 0.28, N = 4 23.07 23.17 1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi
Sockperf This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Ping Pong Auto IBRS Linux 6.2 Git 0.5918 1.1836 1.7754 2.3672 2.959 SE +/- 0.012, N = 5 SE +/- 0.010, N = 5 2.533 2.630 1. (CXX) g++ options: --param -O3 -rdynamic
OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Under Load Auto IBRS Linux 6.2 Git 3 6 9 12 15 SE +/- 0.29, N = 24 SE +/- 0.12, N = 5 11.84 12.69 1. (CXX) g++ options: --param -O3 -rdynamic
Auto IBRS Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.2.0-rc3-phx-tip-x86-cpu (x86_64), Desktop: GNOME Shell 43.1, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2)Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 20 January 2023 06:14 by user phoronix.
Linux 6.2 Git Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 2 x 1920GB SAMSUNG MZWLJ1T9HBJR-00007, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.2.0-060200rc4daily20230121-generic (x86_64), Desktop: GNOME Shell 43.1, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2)Python Notes: Python 3.10.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 21 January 2023 06:43 by user phoronix.