AMD Zen 4 Auto IBRS Performance Initial performance tests of Auto IBRS set for Linux 6.3 compared to prior Retpoline implementation... Benchmarks by Michael Larabel for a future article. More to come.
HTML result view exported from: https://openbenchmarking.org/result/2301191-NE-AMDZEN4AU67&grs&sro .
AMD Zen 4 Auto IBRS Performance Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution Auto IBRS Linux 6.2 Git AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR X670E HERO (0805 BIOS) AMD Device 14d8 32GB Western Digital WD_BLACK SN850X 1000GB + 2000GB AMD Radeon RX 6800 XT 16GB (2575/1000MHz) AMD Navi 21/23 ASUS MG28U Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 Ubuntu 22.10 6.2.0-rc3-phx-tip-x86-cpu (x86_64) GNOME Shell 43.1 X Server + Wayland 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49) 1.3.224 GCC 12.2.0 + Clang 15.0.2-1 ext4 3840x2160 6.2.0-060200rc4daily20230118-generic (x86_64) OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --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 Disk Details - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203 Java Details - OpenJDK Runtime Environment (build 18.0.2-ea+9-Ubuntu-2) Python Details - Python 3.10.7 Security Details - 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
AMD Zen 4 Auto IBRS Performance sockperf: Throughput stress-ng: IO_uring memtier-benchmark: Redis - 50 - 1:1 compilebench: Compile memtier-benchmark: Redis - 100 - 1:1 compilebench: Initial Create sockperf: Latency Ping Pong stress-ng: Context Switching postmark: Disk Transaction Performance pgbench: 100 - 1000 - Read Only - Average Latency pgbench: 100 - 1000 - Read Only nginx: 500 nginx: 200 dragonflydb: 200 - 5:1 renaissance: Finagle HTTP Requests nginx: 1000 stress-ng: NUMA osbench: Create Files nginx: 100 cockroach: KV, 50% Reads - 512 memtier-benchmark: Redis - 50 - 10:1 sockperf: Latency Under Load pgbench: 1 - 1000 - Read Only - Average Latency pgbench: 1 - 1000 - Read Only pgbench: 1 - 800 - Read Only pgbench: 1 - 800 - Read Only - Average Latency osbench: Create Processes selenium: ARES-6 - Firefox redis: GET - 500 wireguard: selenium: WASM collisionDetection - Firefox selenium: ARES-6 - Google Chrome pgbench: 100 - 800 - Read Only redis: SET - 500 selenium: Speedometer - Firefox dragonflydb: 200 - 1:5 stress-ng: Semaphores pgbench: 100 - 800 - Read Only - Average Latency spark: 1000000 - 100 - SHA-512 Benchmark Time renaissance: Genetic Algorithm Using Jenetics + Futures selenium: WASM collisionDetection - Google Chrome brl-cad: VGR Performance Metric node-express-loadtest: cockroach: KV, 50% Reads - 128 memtier-benchmark: Redis - 50 - 1:10 cockroach: KV, 10% Reads - 128 rocksdb: Read While Writing selenium: PSPDFKit WASM - Google Chrome selenium: Jetstream 2 - Firefox spark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframe pgbench: 100 - 1000 - Read Write - Average Latency cockroach: KV, 10% Reads - 512 pgbench: 100 - 1000 - Read Write memtier-benchmark: Redis - 100 - 1:10 selenium: Kraken - Google Chrome git: Time To Complete Common Git Commands build-linux-kernel: defconfig rocksdb: Rand Fill Sync osbench: Create Threads keydb: cockroach: MoVR - 512 cockroach: MoVR - 128 stress-ng: MMAP rocksdb: Update Rand selenium: Kraken - Firefox darktable: Masskrug - CPU-only rocksdb: Seq Fill spark: 1000000 - 100 - Calculate Pi Benchmark rocksdb: Read Rand Write Rand influxdb: 64 - 10000 - 2,5000,1 - 10000 build-godot: Time To Compile rocksdb: Rand Read rocksdb: Rand Fill sqlite-speedtest: Timed Time - Size 1,000 selenium: Speedometer - Google Chrome build-linux-kernel: allmodconfig build-llvm: Ninja dacapobench: Jython darktable: Boat - CPU-only selenium: PSPDFKit WASM - Firefox influxdb: 4 - 10000 - 2,5000,1 - 10000 selenium: WASM imageConvolute - Firefox ctx-clock: Context Switch Time hbase: 10000 - Rand Write - 32 stress-ng: Mutex redis: GET - 50 renaissance: Scala Dotty Auto IBRS Linux 6.2 Git 1266609 32691.97 3960450.20 3675.16 3947399.46 1017.49 1.598 11114905.04 10481 0.537 1863778 179505.24 199332.30 5408933.47 1849.1 154199.72 614.55 9.686768 200874.58 93448.8 3396653.68 9.688 0.496 2015292 2030567 0.394 25.343100 17.60 5331656.2 137.679 240.3 6.8 1880473 4344015.0 297.1 5764962.90 3524027.40 0.426 1.74 957.6 215.17 432531 11903 98611.6 3939082.09 78351.4 4347155 2432 207.767 2.76 23.100 76297.1 43294 3956141.90 346.6 31.271 46.755 39920 9.973049 927313.07 1248.2 1248.7 390.87 1019843 475.4 2.634 1492618 56.03 3379652 2739091.6 51.360 156376252 1456384 35.312 374 527.722 256.854 2132 2.338 2100 2499344.1 17.9 180 443048 14819410.50 5829734.5 437.9 1105464 29637.52 3593245.83 3377.30 3632272.71 948.05 1.690 10603604.01 10050 0.560 1787889 172610.31 192280.90 5221293.14 1914.8 148948.26 593.72 10.017125 195874.71 91139.8 3314366.22 9.927 0.508 1968640 1984551 0.403 25.879542 17.97 5223331.8 140.522 245.1 6.68 1855879 4289025.5 300.8 5694104.51 3481253.23 0.431 1.76 968.4 217.59 427839 11779 97655.5 3901210.72 77695.1 4383598 2452 206.207 2.78 23.260 75773.2 42997 3930034.99 348.9 31.464 47.035 39697 10.027091 922807.85 1242.2 1242.8 389.03 1015269 477.5 2.645 1486446 55.83 3369315 2747445.2 51.204 156849249 1452081 35.409 375 529.023 256.255 2128 2.334 2102 2498029.1 17.9 180 437304 13964278.58 4791960.8 474.3 OpenBenchmarking.org
Sockperf Test: Throughput OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.7 Test: Throughput Auto IBRS Linux 6.2 Git 300K 600K 900K 1200K 1500K SE +/- 5863.99, N = 5 SE +/- 5004.81, N = 5 1266609 1105464 1. (CXX) g++ options: --param -O3 -rdynamic
Stress-NG Test: IO_uring OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: IO_uring Auto IBRS Linux 6.2 Git 7K 14K 21K 28K 35K SE +/- 361.53, N = 15 SE +/- 347.68, N = 4 32691.97 29637.52 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
memtier_benchmark Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:1 OpenBenchmarking.org Ops/sec, More Is Better memtier_benchmark 1.4 Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:1 Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 36099.95, N = 3 SE +/- 33274.79, N = 6 3960450.20 3593245.83 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Compile Bench Test: Compile OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile Auto IBRS Linux 6.2 Git 800 1600 2400 3200 4000 SE +/- 6.63, N = 3 SE +/- 15.37, N = 3 3675.16 3377.30
memtier_benchmark Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:1 OpenBenchmarking.org Ops/sec, More Is Better memtier_benchmark 1.4 Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:1 Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 44405.62, N = 4 SE +/- 28049.24, N = 3 3947399.46 3632272.71 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Compile Bench Test: Initial Create OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create Auto IBRS Linux 6.2 Git 200 400 600 800 1000 SE +/- 1.47, N = 3 SE +/- 9.40, N = 3 1017.49 948.05
Sockperf Test: Latency Ping Pong OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Ping Pong Auto IBRS Linux 6.2 Git 0.3803 0.7606 1.1409 1.5212 1.9015 SE +/- 0.006, N = 5 SE +/- 0.004, N = 5 1.598 1.690 1. (CXX) g++ options: --param -O3 -rdynamic
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Context Switching Auto IBRS Linux 6.2 Git 2M 4M 6M 8M 10M SE +/- 121682.79, N = 3 SE +/- 115542.49, N = 5 11114905.04 10603604.01 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
PostMark Disk Transaction Performance OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance Auto IBRS Linux 6.2 Git 2K 4K 6K 8K 10K SE +/- 72.47, N = 15 SE +/- 80.98, N = 9 10481 10050 1. (CC) gcc options: -O3
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.126 0.252 0.378 0.504 0.63 SE +/- 0.004, N = 3 SE +/- 0.004, N = 10 0.537 0.560 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only Auto IBRS Linux 6.2 Git 400K 800K 1200K 1600K 2000K SE +/- 15499.39, N = 3 SE +/- 14125.73, N = 10 1863778 1787889 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
nginx Connections: 500 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 +/- 51.96, N = 3 SE +/- 307.50, N = 3 179505.24 172610.31 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 Auto IBRS Linux 6.2 Git 40K 80K 120K 160K 200K SE +/- 110.51, N = 3 SE +/- 449.71, N = 3 199332.30 192280.90 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Dragonflydb Clients: 200 - Set To Get Ratio: 5:1 OpenBenchmarking.org Ops/sec, More Is Better Dragonflydb 0.6 Clients: 200 - Set To Get Ratio: 5:1 Auto IBRS Linux 6.2 Git 1.2M 2.4M 3.6M 4.8M 6M SE +/- 25693.70, N = 3 SE +/- 39853.40, N = 3 5408933.47 5221293.14 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Renaissance Test: Finagle HTTP Requests OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Finagle HTTP Requests Auto IBRS Linux 6.2 Git 400 800 1200 1600 2000 SE +/- 20.35, N = 3 SE +/- 16.07, N = 3 1849.1 1914.8 MIN: 1692.3 / MAX: 1870.64 MIN: 1773.42 / MAX: 1939
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 Auto IBRS Linux 6.2 Git 30K 60K 90K 120K 150K SE +/- 195.40, N = 3 SE +/- 360.71, N = 3 154199.72 148948.26 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Stress-NG Test: NUMA OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: NUMA Auto IBRS Linux 6.2 Git 130 260 390 520 650 SE +/- 0.70, N = 3 SE +/- 1.70, N = 3 614.55 593.72 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
OSBench Test: Create Files OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files Auto IBRS Linux 6.2 Git 3 6 9 12 15 SE +/- 0.013312, N = 3 SE +/- 0.024100, N = 3 9.686768 10.017125 1. (CC) gcc options: -lm
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 Auto IBRS Linux 6.2 Git 40K 80K 120K 160K 200K SE +/- 336.57, N = 3 SE +/- 398.86, N = 3 200874.58 195874.71 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
CockroachDB Workload: KV, 50% Reads - Concurrency: 512 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 512 Auto IBRS Linux 6.2 Git 20K 40K 60K 80K 100K SE +/- 1007.53, N = 3 SE +/- 88.42, N = 3 93448.8 91139.8
memtier_benchmark Protocol: Redis - Clients: 50 - Set To Get Ratio: 10:1 OpenBenchmarking.org Ops/sec, More Is Better memtier_benchmark 1.4 Protocol: Redis - Clients: 50 - Set To Get Ratio: 10:1 Auto IBRS Linux 6.2 Git 700K 1400K 2100K 2800K 3500K SE +/- 21454.45, N = 3 SE +/- 28131.12, N = 3 3396653.68 3314366.22 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Sockperf Test: Latency Under Load 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 +/- 0.085, N = 5 SE +/- 0.116, N = 25 9.688 9.927 1. (CXX) g++ options: --param -O3 -rdynamic
PostgreSQL Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.1143 0.2286 0.3429 0.4572 0.5715 SE +/- 0.005, N = 3 SE +/- 0.003, N = 3 0.496 0.508 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 1 - Clients: 1000 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only Auto IBRS Linux 6.2 Git 400K 800K 1200K 1600K 2000K SE +/- 21370.27, N = 3 SE +/- 10622.83, N = 3 2015292 1968640 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 1 - Clients: 800 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Only Auto IBRS Linux 6.2 Git 400K 800K 1200K 1600K 2000K SE +/- 29002.25, N = 3 SE +/- 16388.51, N = 3 2030567 1984551 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
PostgreSQL Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.0907 0.1814 0.2721 0.3628 0.4535 SE +/- 0.006, N = 3 SE +/- 0.003, N = 3 0.394 0.403 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OSBench Test: Create Processes OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes Auto IBRS Linux 6.2 Git 6 12 18 24 30 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 25.34 25.88 1. (CC) gcc options: -lm
Selenium Benchmark: ARES-6 - Browser: Firefox OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Firefox Auto IBRS Linux 6.2 Git 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 17.60 17.97 1. firefox 109.0
Redis Test: GET - Parallel Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better Redis 7.0.4 Test: GET - Parallel Connections: 500 Auto IBRS Linux 6.2 Git 1.1M 2.2M 3.3M 4.4M 5.5M SE +/- 18088.97, N = 3 SE +/- 80377.64, N = 15 5331656.2 5223331.8 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
WireGuard + Linux Networking Stack Stress Test OpenBenchmarking.org Seconds, Fewer Is Better WireGuard + Linux Networking Stack Stress Test Auto IBRS Linux 6.2 Git 30 60 90 120 150 SE +/- 1.22, N = 3 SE +/- 0.87, N = 3 137.68 140.52
Selenium Benchmark: WASM collisionDetection - Browser: Firefox OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM collisionDetection - Browser: Firefox Auto IBRS Linux 6.2 Git 50 100 150 200 250 SE +/- 1.83, N = 3 SE +/- 1.56, N = 3 240.3 245.1 1. firefox 109.0
Selenium Benchmark: ARES-6 - Browser: Google Chrome OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: ARES-6 - Browser: Google Chrome Auto IBRS Linux 6.2 Git 2 4 6 8 10 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 6.80 6.68 1. chrome 109.0.5414.74
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only Auto IBRS Linux 6.2 Git 400K 800K 1200K 1600K 2000K SE +/- 9553.61, N = 3 SE +/- 13559.13, N = 3 1880473 1855879 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Redis Test: SET - Parallel Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better Redis 7.0.4 Test: SET - Parallel Connections: 500 Auto IBRS Linux 6.2 Git 900K 1800K 2700K 3600K 4500K SE +/- 8397.44, N = 3 SE +/- 19184.12, N = 3 4344015.0 4289025.5 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Selenium Benchmark: Speedometer - Browser: Firefox OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Firefox Auto IBRS Linux 6.2 Git 70 140 210 280 350 SE +/- 1.95, N = 3 SE +/- 1.25, N = 3 297.1 300.8 1. firefox 109.0
Dragonflydb Clients: 200 - Set To Get Ratio: 1:5 OpenBenchmarking.org Ops/sec, More Is Better Dragonflydb 0.6 Clients: 200 - Set To Get Ratio: 1:5 Auto IBRS Linux 6.2 Git 1.2M 2.4M 3.6M 4.8M 6M SE +/- 31841.88, N = 3 SE +/- 54495.49, N = 3 5764962.90 5694104.51 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Semaphores Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 1088.41, N = 3 SE +/- 17099.65, N = 3 3524027.40 3481253.23 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
PostgreSQL Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency Auto IBRS Linux 6.2 Git 0.097 0.194 0.291 0.388 0.485 SE +/- 0.002, N = 3 SE +/- 0.003, N = 3 0.426 0.431 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Apache Spark Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time 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 0.396 0.792 1.188 1.584 1.98 SE +/- 0.02, N = 6 SE +/- 0.01, N = 3 1.74 1.76
Renaissance Test: Genetic Algorithm Using Jenetics + Futures OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Genetic Algorithm Using Jenetics + Futures Auto IBRS Linux 6.2 Git 200 400 600 800 1000 SE +/- 6.07, N = 3 SE +/- 8.53, N = 3 957.6 968.4 MIN: 899.95 / MAX: 995.27 MIN: 931.14 / MAX: 997.79
Selenium Benchmark: WASM collisionDetection - Browser: Google Chrome OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM collisionDetection - Browser: Google Chrome Auto IBRS Linux 6.2 Git 50 100 150 200 250 SE +/- 2.35, N = 3 SE +/- 0.33, N = 3 215.17 217.59 1. chrome 109.0.5414.74
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.34 VGR Performance Metric Auto IBRS Linux 6.2 Git 90K 180K 270K 360K 450K 432531 427839 1. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6
Node.js Express HTTP Load Test OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test Auto IBRS Linux 6.2 Git 3K 6K 9K 12K 15K SE +/- 79.93, N = 3 SE +/- 35.97, N = 3 11903 11779
CockroachDB Workload: KV, 50% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 128 Auto IBRS Linux 6.2 Git 20K 40K 60K 80K 100K SE +/- 828.91, N = 3 SE +/- 605.18, N = 3 98611.6 97655.5
memtier_benchmark Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better memtier_benchmark 1.4 Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10 Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 16843.77, N = 3 SE +/- 50455.65, N = 3 3939082.09 3901210.72 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
CockroachDB Workload: KV, 10% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 128 Auto IBRS Linux 6.2 Git 20K 40K 60K 80K 100K SE +/- 561.86, N = 3 SE +/- 597.83, N = 3 78351.4 77695.1
RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Read While Writing Auto IBRS Linux 6.2 Git 900K 1800K 2700K 3600K 4500K SE +/- 9993.64, N = 3 SE +/- 23415.21, N = 3 4347155 4383598 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome Auto IBRS Linux 6.2 Git 500 1000 1500 2000 2500 SE +/- 5.51, N = 3 SE +/- 16.48, N = 3 2432 2452 1. chrome 109.0.5414.74
Selenium Benchmark: Jetstream 2 - Browser: Firefox OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream 2 - Browser: Firefox Auto IBRS Linux 6.2 Git 50 100 150 200 250 SE +/- 1.86, N = 3 207.77 206.21 1. firefox 109.0
Apache Spark Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe Auto IBRS Linux 6.2 Git 0.6255 1.251 1.8765 2.502 3.1275 SE +/- 0.01, N = 6 SE +/- 0.03, N = 3 2.76 2.78
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency Auto IBRS Linux 6.2 Git 6 12 18 24 30 SE +/- 0.14, N = 3 SE +/- 0.17, N = 3 23.10 23.26 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
CockroachDB Workload: KV, 10% Reads - Concurrency: 512 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 512 Auto IBRS Linux 6.2 Git 16K 32K 48K 64K 80K SE +/- 258.53, N = 3 SE +/- 162.67, N = 3 76297.1 75773.2
PostgreSQL Scaling Factor: 100 - Clients: 1000 - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write Auto IBRS Linux 6.2 Git 9K 18K 27K 36K 45K SE +/- 268.00, N = 3 SE +/- 316.61, N = 3 43294 42997 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
memtier_benchmark Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better memtier_benchmark 1.4 Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10 Auto IBRS Linux 6.2 Git 800K 1600K 2400K 3200K 4000K SE +/- 47120.15, N = 4 SE +/- 44656.72, N = 4 3956141.90 3930034.99 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
Selenium Benchmark: Kraken - Browser: Google Chrome OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Google Chrome Auto IBRS Linux 6.2 Git 80 160 240 320 400 SE +/- 0.20, N = 3 SE +/- 0.94, N = 3 346.6 348.9 1. chrome 109.0.5414.74
Git Time To Complete Common Git Commands OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands Auto IBRS Linux 6.2 Git 7 14 21 28 35 SE +/- 0.29, N = 3 SE +/- 0.32, N = 3 31.27 31.46 1. git version 2.37.2
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: defconfig Auto IBRS Linux 6.2 Git 11 22 33 44 55 SE +/- 0.29, N = 3 SE +/- 0.33, N = 3 46.76 47.04
RocksDB Test: Random Fill Sync OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Fill Sync Auto IBRS Linux 6.2 Git 9K 18K 27K 36K 45K SE +/- 26.01, N = 3 SE +/- 33.49, N = 3 39920 39697 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OSBench Test: Create Threads OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads Auto IBRS Linux 6.2 Git 3 6 9 12 15 SE +/- 0.093146, N = 3 SE +/- 0.013857, N = 3 9.973049 10.027091 1. (CC) gcc options: -lm
KeyDB OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.2.0 Auto IBRS Linux 6.2 Git 200K 400K 600K 800K 1000K SE +/- 2161.80, N = 3 SE +/- 1966.20, N = 3 927313.07 922807.85 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
CockroachDB Workload: MoVR - Concurrency: 512 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: MoVR - Concurrency: 512 Auto IBRS Linux 6.2 Git 300 600 900 1200 1500 SE +/- 1.00, N = 3 SE +/- 1.40, N = 3 1248.2 1242.2
CockroachDB Workload: MoVR - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: MoVR - Concurrency: 128 Auto IBRS Linux 6.2 Git 300 600 900 1200 1500 SE +/- 1.95, N = 3 SE +/- 0.78, N = 3 1248.7 1242.8
Stress-NG Test: MMAP OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: MMAP Auto IBRS Linux 6.2 Git 80 160 240 320 400 SE +/- 0.29, N = 3 SE +/- 1.47, N = 3 390.87 389.03 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Update Random Auto IBRS Linux 6.2 Git 200K 400K 600K 800K 1000K SE +/- 3021.46, N = 3 SE +/- 1241.80, N = 3 1019843 1015269 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Selenium Benchmark: Kraken - Browser: Firefox OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: Kraken - Browser: Firefox Auto IBRS Linux 6.2 Git 100 200 300 400 500 SE +/- 1.47, N = 3 SE +/- 1.24, N = 3 475.4 477.5 1. firefox 109.0
Darktable Test: Masskrug - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 4.0.0 Test: Masskrug - Acceleration: CPU-only Auto IBRS Linux 6.2 Git 0.5951 1.1902 1.7853 2.3804 2.9755 SE +/- 0.003, N = 3 SE +/- 0.009, N = 3 2.634 2.645
RocksDB Test: Sequential Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Sequential Fill Auto IBRS Linux 6.2 Git 300K 600K 900K 1200K 1500K SE +/- 3927.45, N = 3 SE +/- 5032.31, N = 3 1492618 1486446 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Apache Spark Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Auto IBRS Linux 6.2 Git 13 26 39 52 65 SE +/- 0.31, N = 6 SE +/- 0.19, N = 3 56.03 55.83
RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Read Random Write Random Auto IBRS Linux 6.2 Git 700K 1400K 2100K 2800K 3500K SE +/- 2704.70, N = 3 SE +/- 7285.88, N = 3 3379652 3369315 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
InfluxDB Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Auto IBRS Linux 6.2 Git 600K 1200K 1800K 2400K 3000K SE +/- 5526.60, N = 3 SE +/- 10682.61, N = 3 2739091.6 2747445.2
Timed Godot Game Engine Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile Auto IBRS Linux 6.2 Git 12 24 36 48 60 SE +/- 0.12, N = 3 SE +/- 0.14, N = 3 51.36 51.20
RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Read Auto IBRS Linux 6.2 Git 30M 60M 90M 120M 150M SE +/- 417674.12, N = 3 SE +/- 742538.53, N = 3 156376252 156849249 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
RocksDB Test: Random Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Fill Auto IBRS Linux 6.2 Git 300K 600K 900K 1200K 1500K SE +/- 4987.59, N = 3 SE +/- 1973.97, N = 3 1456384 1452081 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
SQLite Speedtest Timed Time - Size 1,000 OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 Auto IBRS Linux 6.2 Git 8 16 24 32 40 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 35.31 35.41 1. (CC) gcc options: -O2 -lz
Selenium Benchmark: Speedometer - Browser: Google Chrome OpenBenchmarking.org Runs Per Minute, More Is Better Selenium Benchmark: Speedometer - Browser: Google Chrome Auto IBRS Linux 6.2 Git 80 160 240 320 400 SE +/- 2.96, N = 3 SE +/- 2.03, N = 3 374 375 1. chrome 109.0.5414.74
Timed Linux Kernel Compilation Build: allmodconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: allmodconfig Auto IBRS Linux 6.2 Git 110 220 330 440 550 SE +/- 1.07, N = 3 SE +/- 0.55, N = 3 527.72 529.02
Timed LLVM Compilation Build System: Ninja OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 13.0 Build System: Ninja Auto IBRS Linux 6.2 Git 60 120 180 240 300 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 256.85 256.26
DaCapo Benchmark Java Test: Jython OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython Auto IBRS Linux 6.2 Git 500 1000 1500 2000 2500 SE +/- 9.91, N = 4 SE +/- 15.71, N = 4 2132 2128
Darktable Test: Boat - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 4.0.0 Test: Boat - Acceleration: CPU-only Auto IBRS Linux 6.2 Git 0.5261 1.0522 1.5783 2.1044 2.6305 SE +/- 0.005, N = 3 SE +/- 0.011, N = 3 2.338 2.334
Selenium Benchmark: PSPDFKit WASM - Browser: Firefox OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox Auto IBRS Linux 6.2 Git 500 1000 1500 2000 2500 SE +/- 5.81, N = 3 SE +/- 4.37, N = 3 2100 2102 1. firefox 109.0
InfluxDB Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Auto IBRS Linux 6.2 Git 500K 1000K 1500K 2000K 2500K SE +/- 7750.93, N = 3 SE +/- 4064.43, N = 3 2499344.1 2498029.1
Selenium Benchmark: WASM imageConvolute - Browser: Firefox OpenBenchmarking.org ms, Fewer Is Better Selenium Benchmark: WASM imageConvolute - Browser: Firefox Auto IBRS Linux 6.2 Git 4 8 12 16 20 SE +/- 0.09, N = 3 SE +/- 0.13, N = 3 17.9 17.9 1. firefox 109.0
ctx_clock Context Switch Time OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time Auto IBRS Linux 6.2 Git 40 80 120 160 200 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 180 180
Apache HBase Rows: 10000 - Test: Random Write - Clients: 32 OpenBenchmarking.org Rows Per Second, More Is Better Apache HBase 2.5.0 Rows: 10000 - Test: Random Write - Clients: 32 Auto IBRS Linux 6.2 Git 90K 180K 270K 360K 450K SE +/- 12140.09, N = 15 SE +/- 12894.69, N = 15 443048 437304
Stress-NG Test: Mutex OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.14.06 Test: Mutex Auto IBRS Linux 6.2 Git 3M 6M 9M 12M 15M SE +/- 378248.33, N = 15 SE +/- 78291.25, N = 3 14819410.50 13964278.58 1. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmd -lrt -lsctp -lz -pthread
Redis Test: GET - Parallel Connections: 50 OpenBenchmarking.org Requests Per Second, More Is Better Redis 7.0.4 Test: GET - Parallel Connections: 50 Auto IBRS Linux 6.2 Git 1.2M 2.4M 3.6M 4.8M 6M SE +/- 81129.32, N = 15 SE +/- 87118.74, N = 15 5829734.5 4791960.8 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.14 Test: Scala Dotty Auto IBRS Linux 6.2 Git 100 200 300 400 500 SE +/- 8.70, N = 15 SE +/- 4.22, N = 3 437.9 474.3 MIN: 329.62 / MAX: 645.18 MIN: 339.98 / MAX: 623.77
Phoronix Test Suite v10.8.4