Linux 5.9 Test With page_lock_unfairness Benchmarks by Michael Larabel testing of 14 September page_lock_unfairness patch. Kconfig and other bits same throughout kernels, PLU patches done with Linus' patches from Monday.
HTML result view exported from: https://openbenchmarking.org/result/2009154-FI-LINUX58CO57&sor&grs .
Linux 5.9 Test With page_lock_unfairness Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution v5.8 v5.9 Git Default PLU 1000 PLU 0 PLU 1 PLU 2 PLU 3 PLU 4 PLU 5 AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads) MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS) AMD Starship/Matisse 32GB 1000GB Sabrent Rocket 4.0 1TB Sapphire AMD Radeon RX 5500/5500M / Pro 5500M 4GB AMD Navi 10 HDMI Audio ASUS MG28U Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 20.04 5.8.0-050800-generic (x86_64) GNOME Shell 3.36.4 X Server 1.20.8 modesetting 1.20.8 4.6 Mesa 20.0.8 (LLVM 10.0.0) GCC 9.3.0 ext4 3840x2160 5.9.0-rc4-pts (x86_64) 20200910 5.9.0-rc5-14sep-patch (x86_64) 20200914 OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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 Processor Details - Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Linux 5.9 Test With page_lock_unfairness apache-siege: 200 apache-siege: 100 sqlite: 8 hackbench: 4 - Process apache-siege: 50 hackbench: 4 - Thread hackbench: 2 - Process hackbench: 2 - Thread fio: Rand Read - Linux AIO - Yes - No - 4KB - Default Test Directory fio: Seq Read - Linux AIO - Yes - No - 2MB - Default Test Directory fio: Seq Read - Linux AIO - Yes - No - 2MB - Default Test Directory fio: Rand Read - Linux AIO - Yes - No - 4KB - Default Test Directory redis: GET hackbench: 1 - Process hackbench: 1 - Thread redis: SET redis: SADD nginx: Static Web Page Serving redis: LPUSH apache-siege: 1 fio: Rand Write - Linux AIO - Yes - No - 4KB - Default Test Directory fio: Rand Write - Linux AIO - Yes - No - 4KB - Default Test Directory apache-siege: 250 redis: LPOP fio: Seq Write - Linux AIO - Yes - No - 2MB - Default Test Directory fio: Seq Write - Linux AIO - Yes - No - 2MB - Default Test Directory v5.8 v5.9 Git Default PLU 1000 PLU 0 PLU 1 PLU 2 PLU 3 PLU 4 PLU 5 47825.12 51255.73 50.581 5.265 39280.51 5.660 3.642 3.788 15800 760 1528 61.8 2656138.92 2.760 2.945 1989099.04 2095660.42 42691.90 1602867.64 7684.81 1173 300000 55279.65 2582974.69 1159 2326 20174.78 21926.62 46.483 6.729 28240.71 7.106 4.515 4.761 15867 837 1681 62.1 2111756.79 3.212 3.366 1723206.54 1962098.59 39040.06 1489837.96 7390.86 1141 292000 20282.62 1523888.21 1077 2162 48190.05 42001.86 47.602 6.652 39708.30 7.144 4.533 4.739 11367 755 1518 44.4 2194355.4 3.104 3.324 1820617.25 1974406.96 40274.52 1527638.92 7579.49 1144 293000 67639.46 1540468.42 1107 2222 20095.10 21528.43 47.530 6.732 26645.15 7.167 4.511 4.764 15900 945 1899 62.1 2217841.7 3.208 3.409 1772165.71 2013620.17 39963.52 1512885.79 7937.84 1138 291333 20181.98 1576775.71 1183 2372 48524.44 37138.49 47.583 8.205 38709.95 8.580 5.206 5.371 15900 1068 2143 62.2 2178989.8 3.363 3.591 1745422.5 1965123.21 40553.38 1519532.71 7464.61 1125 288000 40505.37 1535369.42 966 1939 47823.09 38086.58 79.370 8.192 39712.82 8.603 5.219 5.367 11300 846 1700 44.3 2217654.08 3.374 3.579 1758507.92 1912228.67 40676.68 1527284.96 7552.61 1131 289667 56914.07 2088864.07 959 1926 47751.02 38057.72 84.624 6.674 41959.67 7.098 4.537 4.744 16000 1067 2142 62.5 2169197.5 3.155 3.265 1766810.25 1938042.50 40575.55 1506277.25 7475.96 1133 289667 55285.35 2130440.16 841 1690 68286.02 56350.51 47.67 6.723 38870.90 7.103 4.511 4.732 16000 861 1730 62.5 2206395.33 3.189 3.395 1780663.08 1967501.79 40266.94 1516495.37 7638.69 1122 287000 55499.25 1512401.66 1102 2212 47662.08 37868.57 47.725 8.292 41301.97 8.571 5.227 5.383 15867 1059 2126 62.0 2158856.83 3.397 3.509 1775156.63 1968609.69 40337.05 1514181.04 7735.75 1144 293000 74347.77 1553886.50 955 1918 OpenBenchmarking.org
Apache Siege Concurrent Users: 200 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 PLU 4 PLU 1 Default PLU 1000 v5.8 PLU 2 PLU 3 PLU 5 v5.9 Git PLU 0 15K 30K 45K 60K 75K SE +/- 1783.44, N = 15 SE +/- 755.92, N = 3 SE +/- 551.40, N = 3 SE +/- 162.84, N = 3 SE +/- 584.45, N = 3 SE +/- 506.78, N = 3 SE +/- 184.37, N = 3 SE +/- 141.27, N = 3 SE +/- 81.29, N = 3 68286.02 48524.44 48190.05 47825.12 47823.09 47751.02 47662.08 20174.78 20095.10 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 100 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 PLU 4 v5.8 Default PLU 1000 PLU 2 PLU 3 PLU 5 PLU 1 v5.9 Git PLU 0 12K 24K 36K 48K 60K SE +/- 940.16, N = 15 SE +/- 786.44, N = 15 SE +/- 238.29, N = 3 SE +/- 138.58, N = 3 SE +/- 335.61, N = 3 SE +/- 278.88, N = 3 SE +/- 104.00, N = 3 SE +/- 304.87, N = 4 SE +/- 263.19, N = 6 56350.51 51255.73 42001.86 38086.58 38057.72 37868.57 37138.49 21926.62 21528.43 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
SQLite Threads / Copies: 8 OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.30.1 Threads / Copies: 8 v5.9 Git PLU 0 PLU 1 Default PLU 1000 PLU 4 PLU 5 v5.8 PLU 2 PLU 3 20 40 60 80 100 SE +/- 0.18, N = 3 SE +/- 0.30, N = 3 SE +/- 0.16, N = 3 SE +/- 0.17, N = 3 SE +/- 0.58, N = 3 SE +/- 0.14, N = 3 SE +/- 0.07, N = 3 SE +/- 8.24, N = 12 SE +/- 5.48, N = 15 46.48 47.53 47.58 47.60 47.67 47.73 50.58 79.37 84.62 1. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
Hackbench Count: 4 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process v5.8 Default PLU 1000 PLU 3 PLU 4 v5.9 Git PLU 0 PLU 2 PLU 1 PLU 5 2 4 6 8 10 SE +/- 0.076, N = 4 SE +/- 0.043, N = 3 SE +/- 0.058, N = 3 SE +/- 0.027, N = 3 SE +/- 0.033, N = 3 SE +/- 0.074, N = 3 SE +/- 0.014, N = 3 SE +/- 0.037, N = 3 SE +/- 0.029, N = 3 5.265 6.652 6.674 6.723 6.729 6.732 8.192 8.205 8.292 1. (CC) gcc options: -lpthread
Apache Siege Concurrent Users: 50 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 PLU 3 PLU 5 PLU 2 Default PLU 1000 v5.8 PLU 4 PLU 1 v5.9 Git PLU 0 9K 18K 27K 36K 45K SE +/- 184.77, N = 3 SE +/- 492.59, N = 6 SE +/- 630.82, N = 3 SE +/- 554.30, N = 3 SE +/- 250.75, N = 3 SE +/- 118.47, N = 3 SE +/- 72.14, N = 3 SE +/- 359.10, N = 5 SE +/- 301.81, N = 6 41959.67 41301.97 39712.82 39708.30 39280.51 38870.90 38709.95 28240.71 26645.15 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Hackbench Count: 4 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread v5.8 PLU 3 PLU 4 v5.9 Git Default PLU 1000 PLU 0 PLU 5 PLU 1 PLU 2 2 4 6 8 10 SE +/- 0.054, N = 3 SE +/- 0.034, N = 3 SE +/- 0.041, N = 3 SE +/- 0.027, N = 3 SE +/- 0.019, N = 3 SE +/- 0.029, N = 3 SE +/- 0.046, N = 3 SE +/- 0.012, N = 3 SE +/- 0.051, N = 3 5.660 7.098 7.103 7.106 7.144 7.167 8.571 8.580 8.603 1. (CC) gcc options: -lpthread
Hackbench Count: 2 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process v5.8 PLU 0 PLU 4 v5.9 Git Default PLU 1000 PLU 3 PLU 1 PLU 2 PLU 5 1.1761 2.3522 3.5283 4.7044 5.8805 SE +/- 0.030, N = 3 SE +/- 0.035, N = 3 SE +/- 0.032, N = 3 SE +/- 0.011, N = 3 SE +/- 0.019, N = 3 SE +/- 0.023, N = 3 SE +/- 0.019, N = 3 SE +/- 0.009, N = 3 SE +/- 0.023, N = 3 3.642 4.511 4.511 4.515 4.533 4.537 5.206 5.219 5.227 1. (CC) gcc options: -lpthread
Hackbench Count: 2 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Thread v5.8 PLU 4 Default PLU 1000 PLU 3 v5.9 Git PLU 0 PLU 2 PLU 1 PLU 5 1.2112 2.4224 3.6336 4.8448 6.056 SE +/- 0.037, N = 3 SE +/- 0.029, N = 3 SE +/- 0.011, N = 3 SE +/- 0.019, N = 3 SE +/- 0.009, N = 3 SE +/- 0.008, N = 3 SE +/- 0.031, N = 3 SE +/- 0.013, N = 3 SE +/- 0.021, N = 3 3.788 4.732 4.739 4.744 4.761 4.764 5.367 5.371 5.383 1. (CC) gcc options: -lpthread
Flexible IO Tester Type: Random Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.18 Type: Random Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory PLU 4 PLU 3 PLU 1 PLU 0 PLU 5 v5.9 Git v5.8 Default PLU 1000 PLU 2 3K 6K 9K 12K 15K SE +/- 100.00, N = 3 SE +/- 100.00, N = 3 SE +/- 33.33, N = 3 SE +/- 33.33, N = 3 SE +/- 33.33, N = 3 16000 16000 15900 15900 15867 15867 15800 11367 11300 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Flexible IO Tester Type: Sequential Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.18 Type: Sequential Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory PLU 1 PLU 3 PLU 5 PLU 0 PLU 4 PLU 2 v5.9 Git v5.8 Default PLU 1000 200 400 600 800 1000 SE +/- 1.00, N = 3 SE +/- 3.18, N = 3 SE +/- 1.53, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 SE +/- 2.91, N = 3 1068 1067 1059 945 861 846 837 760 755 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Flexible IO Tester Type: Sequential Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.18 Type: Sequential Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory PLU 1 PLU 3 PLU 5 PLU 0 PLU 4 PLU 2 v5.9 Git v5.8 Default PLU 1000 500 1000 1500 2000 2500 SE +/- 2.52, N = 3 SE +/- 6.66, N = 3 SE +/- 2.96, N = 3 SE +/- 4.06, N = 3 SE +/- 1.73, N = 3 SE +/- 1.67, N = 3 SE +/- 5.78, N = 3 SE +/- 1.67, N = 3 2143 2142 2126 1899 1730 1700 1681 1528 1518 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Flexible IO Tester Type: Random Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.18 Type: Random Read - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory PLU 4 PLU 3 PLU 1 PLU 0 v5.9 Git PLU 5 v5.8 Default PLU 1000 PLU 2 14 28 42 56 70 SE +/- 0.48, N = 3 SE +/- 0.37, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 62.5 62.5 62.2 62.1 62.1 62.0 61.8 44.4 44.3 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: GET v5.8 PLU 0 PLU 2 PLU 4 Default PLU 1000 PLU 1 PLU 3 PLU 5 v5.9 Git 600K 1200K 1800K 2400K 3000K SE +/- 41211.45, N = 3 SE +/- 24494.66, N = 3 SE +/- 20036.24, N = 3 SE +/- 23678.74, N = 3 SE +/- 31523.21, N = 4 SE +/- 19341.01, N = 3 SE +/- 0.00, N = 3 SE +/- 25197.60, N = 3 SE +/- 24495.13, N = 3 2656138.92 2217841.70 2217654.08 2206395.33 2194355.40 2178989.80 2169197.50 2158856.83 2111756.79 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Hackbench Count: 1 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Process v5.8 Default PLU 1000 PLU 3 PLU 4 PLU 0 v5.9 Git PLU 1 PLU 2 PLU 5 0.7643 1.5286 2.2929 3.0572 3.8215 SE +/- 0.038, N = 4 SE +/- 0.039, N = 3 SE +/- 0.052, N = 3 SE +/- 0.017, N = 3 SE +/- 0.023, N = 3 SE +/- 0.016, N = 3 SE +/- 0.048, N = 4 SE +/- 0.042, N = 3 SE +/- 0.032, N = 3 2.760 3.104 3.155 3.189 3.208 3.212 3.363 3.374 3.397 1. (CC) gcc options: -lpthread
Hackbench Count: 1 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Thread v5.8 PLU 3 Default PLU 1000 v5.9 Git PLU 4 PLU 0 PLU 5 PLU 2 PLU 1 0.808 1.616 2.424 3.232 4.04 SE +/- 0.045, N = 3 SE +/- 0.040, N = 3 SE +/- 0.024, N = 3 SE +/- 0.010, N = 3 SE +/- 0.015, N = 3 SE +/- 0.047, N = 3 SE +/- 0.053, N = 3 SE +/- 0.034, N = 3 SE +/- 0.028, N = 3 2.945 3.265 3.324 3.366 3.395 3.409 3.509 3.579 3.591 1. (CC) gcc options: -lpthread
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SET v5.8 Default PLU 1000 PLU 4 PLU 5 PLU 0 PLU 3 PLU 2 PLU 1 v5.9 Git 400K 800K 1200K 1600K 2000K SE +/- 31976.02, N = 3 SE +/- 14383.26, N = 3 SE +/- 14792.31, N = 3 SE +/- 2776.74, N = 3 SE +/- 11992.30, N = 3 SE +/- 4779.13, N = 3 SE +/- 2729.49, N = 3 SE +/- 13866.13, N = 3 SE +/- 7115.44, N = 3 1989099.04 1820617.25 1780663.08 1775156.63 1772165.71 1766810.25 1758507.92 1745422.50 1723206.54 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SADD v5.8 PLU 0 Default PLU 1000 PLU 5 PLU 4 PLU 1 v5.9 Git PLU 3 PLU 2 400K 800K 1200K 1600K 2000K SE +/- 35073.13, N = 15 SE +/- 14082.24, N = 3 SE +/- 26583.73, N = 3 SE +/- 26201.01, N = 4 SE +/- 16787.53, N = 3 SE +/- 21780.46, N = 3 SE +/- 5599.04, N = 3 SE +/- 7482.75, N = 3 SE +/- 13251.53, N = 3 2095660.42 2013620.17 1974406.96 1968609.69 1967501.79 1965123.21 1962098.59 1938042.50 1912228.67 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
NGINX Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving v5.8 PLU 2 PLU 3 PLU 1 PLU 5 Default PLU 1000 PLU 4 PLU 0 v5.9 Git 9K 18K 27K 36K 45K SE +/- 68.19, N = 3 SE +/- 512.74, N = 3 SE +/- 294.56, N = 3 SE +/- 138.82, N = 3 SE +/- 249.35, N = 3 SE +/- 271.04, N = 3 SE +/- 107.21, N = 3 SE +/- 356.07, N = 3 SE +/- 126.49, N = 3 42691.90 40676.68 40575.55 40553.38 40337.05 40274.52 40266.94 39963.52 39040.06 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: LPUSH v5.8 Default PLU 1000 PLU 2 PLU 1 PLU 4 PLU 5 PLU 0 PLU 3 v5.9 Git 300K 600K 900K 1200K 1500K SE +/- 25402.74, N = 15 SE +/- 10515.99, N = 3 SE +/- 20906.91, N = 3 SE +/- 20225.81, N = 3 SE +/- 20795.76, N = 3 SE +/- 20527.26, N = 3 SE +/- 24428.16, N = 3 SE +/- 13754.66, N = 3 SE +/- 13985.92, N = 3 1602867.64 1527638.92 1527284.96 1519532.71 1516495.37 1514181.04 1512885.79 1506277.25 1489837.96 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Apache Siege Concurrent Users: 1 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 1 PLU 0 PLU 5 v5.8 PLU 4 Default PLU 1000 PLU 2 PLU 3 PLU 1 v5.9 Git 2K 4K 6K 8K 10K SE +/- 72.75, N = 3 SE +/- 120.85, N = 3 SE +/- 68.63, N = 15 SE +/- 99.83, N = 5 SE +/- 118.32, N = 3 SE +/- 63.44, N = 15 SE +/- 81.44, N = 7 SE +/- 84.29, N = 3 SE +/- 79.07, N = 3 7937.84 7735.75 7684.81 7638.69 7579.49 7552.61 7475.96 7464.61 7390.86 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Flexible IO Tester Type: Random Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.18 Type: Random Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory v5.8 PLU 5 Default PLU 1000 v5.9 Git PLU 0 PLU 3 PLU 2 PLU 1 PLU 4 300 600 900 1200 1500 SE +/- 15.81, N = 3 SE +/- 9.35, N = 3 SE +/- 8.39, N = 3 SE +/- 11.35, N = 3 SE +/- 17.80, N = 3 SE +/- 6.17, N = 3 SE +/- 16.56, N = 3 SE +/- 9.24, N = 3 SE +/- 15.84, N = 3 1173 1144 1144 1141 1138 1133 1131 1125 1122 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Flexible IO Tester Type: Random Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.18 Type: Random Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory v5.8 PLU 5 Default PLU 1000 v5.9 Git PLU 0 PLU 3 PLU 2 PLU 1 PLU 4 60K 120K 180K 240K 300K SE +/- 4163.33, N = 3 SE +/- 2516.61, N = 3 SE +/- 2309.40, N = 3 SE +/- 3055.05, N = 3 SE +/- 4484.54, N = 3 SE +/- 1666.67, N = 3 SE +/- 4176.65, N = 3 SE +/- 2309.40, N = 3 SE +/- 4000.00, N = 3 300000 293000 293000 292000 291333 289667 289667 288000 287000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Apache Siege Concurrent Users: 250 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 PLU 5 Default PLU 1000 PLU 2 PLU 4 PLU 3 v5.8 PLU 1 v5.9 Git PLU 0 16K 32K 48K 64K 80K SE +/- 1206.46, N = 14 SE +/- 3661.08, N = 12 SE +/- 803.33, N = 3 SE +/- 385.12, N = 3 SE +/- 749.49, N = 3 SE +/- 263.70, N = 3 SE +/- 3017.22, N = 12 SE +/- 118.94, N = 3 SE +/- 168.51, N = 3 74347.77 67639.46 56914.07 55499.25 55285.35 55279.65 40505.37 20282.62 20181.98 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: LPOP v5.8 PLU 3 PLU 2 PLU 0 PLU 5 Default PLU 1000 PLU 1 v5.9 Git PLU 4 600K 1200K 1800K 2400K 3000K SE +/- 63136.63, N = 12 SE +/- 109453.34, N = 12 SE +/- 123677.08, N = 12 SE +/- 15835.35, N = 3 SE +/- 14982.36, N = 3 SE +/- 20011.91, N = 5 SE +/- 25335.54, N = 3 SE +/- 14468.35, N = 3 SE +/- 15179.54, N = 3 2582974.69 2130440.16 2088864.07 1576775.71 1553886.50 1540468.42 1535369.42 1523888.21 1512401.66 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Flexible IO Tester Type: Sequential Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.18 Type: Sequential Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory PLU 0 v5.8 Default PLU 1000 PLU 4 v5.9 Git PLU 1 PLU 2 PLU 5 PLU 3 300 600 900 1200 1500 SE +/- 6.66, N = 3 SE +/- 27.31, N = 12 SE +/- 33.50, N = 15 SE +/- 35.14, N = 12 SE +/- 30.08, N = 15 SE +/- 33.13, N = 13 SE +/- 40.60, N = 15 SE +/- 36.11, N = 13 SE +/- 62.04, N = 15 1183 1159 1107 1102 1077 966 959 955 841 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Flexible IO Tester Type: Sequential Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.18 Type: Sequential Write - Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory PLU 0 v5.8 Default PLU 1000 PLU 4 v5.9 Git PLU 1 PLU 2 PLU 5 PLU 3 500 1000 1500 2000 2500 SE +/- 13.64, N = 3 SE +/- 54.55, N = 12 SE +/- 67.07, N = 15 SE +/- 70.24, N = 12 SE +/- 60.14, N = 15 SE +/- 66.32, N = 13 SE +/- 81.18, N = 15 SE +/- 72.16, N = 13 SE +/- 124.13, N = 15 2372 2326 2222 2212 2162 1939 1926 1918 1690 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl
Phoronix Test Suite v10.8.5