Server Mitigations Ubuntu 20.04 Reference Tests for a future article by Michael Larabel.
Toggling mitigations on/off via "mitigations=" Linux kernel flag for out-of-the-box versus capable mitigations disabled.
Xeon Gold 5220R has hardware mitigations present. AMD EPYC Rome run also included.
HTML result view exported from: https://openbenchmarking.org/result/2004064-NI-SERVERS2677 .
Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations Intel Xeon E5-2687W v3 @ 3.50GHz (10 Cores / 20 Threads) MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS) Intel Xeon E7 v3/Xeon 32GB 80GB INTEL SSDSCKGW08 NVIDIA GeForce GTX 770 2GB Realtek ALC892 G237HL Intel I218-V Ubuntu 20.04 5.4.0-21-generic (x86_64) GNOME Shell 3.36.0 X Server 1.20.7 modesetting 1.20.7 4.3 Mesa 20.0.2 GCC 9.3.0 ext4 1920x1080 NVIDIA NVE4 2GB Intel Xeon E3-1275 v6 @ 4.20GHz (4 Cores / 8 Threads) ASUS P10S-M WS (4401 BIOS) Intel Xeon E3-1200 v6/7th 16GB Samsung SSD 970 EVO Plus 500GB Intel HD P630 3GB (1150MHz) Realtek ALC1150 DELL S2409W 2 x Intel I210 4.6 Mesa 20.0.2 2 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads) TYAN S7106 (V2.01.B40 BIOS) Intel Sky Lake-E DMI3 Registers 94GB 500GB Samsung SSD 860 ASPEED VE228 2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE llvmpipe 93GB 3.3 Mesa 20.0.2 (LLVM 9.0.1 256 bits) 2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads) AMD EPYC 7502P 32-Core @ 2.50GHz (32 Cores / 64 Threads) ASRockRack EPYCD8 (P2.10 BIOS) AMD Starship/Matisse 126GB 280GB INTEL SSDPED1D280GA ASPEED AMD Starship/Matisse 2 x Intel I350 llvmpipe 126GB 3.3 Mesa 20.0.2 (LLVM 9.0.1 128 bits) 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 - E5-2687W v3: Defaults: MQ-DEADLINE / errors=remount-ro,relatime,rw - E5-2687W v3: No Mitigations: MQ-DEADLINE / errors=remount-ro,relatime,rw - E3-1275 v6: Defaults: NONE / errors=remount-ro,relatime,rw - E3-1275 v6: No Mitigations: NONE / errors=remount-ro,relatime,rw - 2 x Gold 6138: Defaults: MQ-DEADLINE / errors=remount-ro,relatime,rw - 2 x Gold 6138: No Mitigations: MQ-DEADLINE / errors=remount-ro,relatime,rw - 2 x Gold 5220R: Defaults: MQ-DEADLINE / errors=remount-ro,relatime,rw - 2 x Gold 5220R: No Mitigations: MQ-DEADLINE / errors=remount-ro,relatime,rw - EPYC 7502P: Defaults: NONE / errors=remount-ro,relatime,rw - EPYC 7502P: No Mitigations: NONE / errors=remount-ro,relatime,rw Processor Details - E5-2687W v3: Defaults: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x43 - E5-2687W v3: No Mitigations: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x43 - E3-1275 v6: Defaults: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xca - E3-1275 v6: No Mitigations: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xca - 2 x Gold 6138: Defaults: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x2000064 - 2 x Gold 6138: No Mitigations: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x2000064 - 2 x Gold 5220R: Defaults: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c - 2 x Gold 5220R: No Mitigations: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c - EPYC 7502P: Defaults: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x830101c - EPYC 7502P: No Mitigations: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x830101c Java Details - OpenJDK Runtime Environment (build 11.0.7-ea+9-post-Ubuntu-1ubuntu1) Python Details - + Python 3.8.2 Security Details - E5-2687W v3: Defaults: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + tsx_async_abort: Not affected - E5-2687W v3: No Mitigations: itlb_multihit: KVM: Vulnerable + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled + tsx_async_abort: Not affected - E3-1275 v6: Defaults: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable - E3-1275 v6: No Mitigations: itlb_multihit: KVM: Vulnerable + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled + tsx_async_abort: Vulnerable - 2 x Gold 6138: Defaults: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + tsx_async_abort: Mitigation of Clear buffers; SMT vulnerable - 2 x Gold 6138: No Mitigations: itlb_multihit: KVM: Vulnerable + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled + tsx_async_abort: Vulnerable - 2 x Gold 5220R: Defaults: itlb_multihit: KVM: Mitigation of Split huge pages + 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 Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled - 2 x Gold 5220R: No Mitigations: itlb_multihit: KVM: Vulnerable + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled + tsx_async_abort: Mitigation of TSX disabled - EPYC 7502P: Defaults: 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 IBRS_FW STIBP: always-on RSB filling + tsx_async_abort: Not affected - EPYC 7502P: No Mitigations: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled + tsx_async_abort: Not affected
blogbench: Read sockperf: Throughput sockperf: Latency Ping Pong iperf: 5201 - 30 Seconds - TCP - 1 iperf: 5201 - 30 Seconds - TCP - 32 dacapobench: Jython dacapobench: Tradesoap renaissance: Rand Forest renaissance: Apache Spark ALS renaissance: Apache Spark PageRank renaissance: Twitter HTTP Requests renaissance: In-Memory Database Shootout renaissance: Akka Unbalanced Cobwebbed Tree renaissance: Genetic Algorithm Using Jenetics + Futures build-linux-kernel: Time To Compile build-llvm: Time To Compile build-php: Time To Compile build2: Time To Compile perl-benchmark: Pod2html perl-benchmark: Interpreter keydb: pgbench: Buffer Test - Normal Load - Read Only pgbench: Buffer Test - Heavy Contention - Read Only sqlite-speedtest: Timed Time - Size 1,000 redis: LPUSH redis: SET rocksdb: Rand Fill rocksdb: Seq Fill mcperf: Get - 16 mcperf: Append - 16 mcperf: Delete - 16 mcperf: Replace - 16 nginx: Static Web Page Serving hbase: Increment - 4 hbase: Increment - 4 hbase: Rand Read - 4 hbase: Rand Read - 4 hbase: Seq Read - 4 hbase: Seq Read - 4 hbase: Async Rand Write - 4 hbase: Async Rand Write - 4 apache: Static Web Page Serving E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations Defaults No Mitigations 1830416 266654 4.807 34129 38523 5411 5079 5619.447 6449.645 24481.436 5838.764 7368.545 13778.505 4769.196 107.825 812.198 82.492 142.671 0.15528423 0.00284140 448533.08 162112.134818 165170.199178 88.927 1280493.84 1550985.79 518865 694894 88133.1 37763.5 84443.9 36941.2 23945.41 8811 452 18668 212 19519 207 5736 695 16731.18 1906597 340151 4.030 35326 36871 5371 4919 5695.296 6468.937 24696.023 5725.518 6856.779 13307.739 4674.329 105.274 809.017 80.833 140.586 0.14966118 0.00247743 561921.38 176460.211767 178879.321153 73.620 1355759.00 1605077.06 678788 1038858 114148.3 49633.8 112817.5 50251.4 31796.96 10138 393 21439 185 22439 180 6651 600 22745.66 647728 486761 3.398 63586 68520 4022 6572 4882.482 6049.134 23626.664 5189.184 5182.954 11118.967 1677.188 176.992 1367.412 99.772 223.760 0.11922478 0.00099263 332575.74 106967.306636 110061.942886 68.143 1709562.63 2036733.29 774300 1003248 122674.1 75114.2 119621.9 75108.9 37413.38 23474 169 41639 95 29375 135 14693 271 27301.81 644528 619109 2.691 70194 72902 3928 6361 4730.423 5893.831 23339.004 5027.579 5007.481 10751.788 1645.886 172.548 1348.513 98.023 219.180 0.11721965 0.00093284 460535.14 121451.018355 121036.732225 57.373 1791215.29 2175635.00 885609 1113455 163813.9 100971.3 163124.0 101249.7 46860.12 26591 149 47925 82 36949 108 16665 241 35270.52 2038525 357391 4.888 34314 33418 4490 6799 5742.317 7143.696 25666.985 7845.882 16087.978 26470.854 6840.662 41.191 304.450 50.958 80.039 0.12653994 0.00375265 365949.97 529354.416803 562889.815935 77.434 1321227.17 1567680.21 141751 142824 88890.2 56358.7 53759.9 36701.9 27430.11 10347 384 18054 219 17630 227 6111 652 20351.02 2129085 417357 3.984 35924 32464 4390 6746 5703.445 6983.425 25213.781 7749.772 14649.671 25007.433 6553.619 40.044 300.307 50.216 78.662 0.12412203 0.00330160 407888.25 564233.096037 598979.964631 66.462 1414137.39 1607048.83 176835 177505 108534.1 67756.4 57494.3 68552.6 34006.44 12134 328 21613 183 22683 178 7938 502 24738.87 2215082 398750 3.941 36719 33326 4824 6701 5593.392 6961.553 24250.424 7163.349 14068.526 23067.161 6606.997 37.478 284.994 48.744 75.463 0.11832707 0.00329017 416533.95 603798.289346 626099.893434 68.258 1371743.83 1706887.67 179667 180768 101109.8 30394.7 107712.7 29106.3 31476.33 12004 332 21155 187 21557 187 7480 533 24204.88 2238187 405922 3.739 36951 33593 4841 6543 5685.463 6965.359 24532.442 7193.924 14030.638 22828.077 6535.201 37.450 283.596 48.313 75.712 0.11850206 0.00325939 418381.24 596866.113606 621603.189448 68.249 1422514.42 1665880.83 182743 184392 117411.7 61641.5 116085.7 68576.5 34256.90 12059 330 21377 184 22365 180 7738 515 24785.26 1839796 449645 4.810 41194 39761 5183 4443 5037.821 6279.249 21806.948 6472.957 8650.313 16488.147 3056.516 37.657 288.263 46.629 72.695 0.14704507 0.00203798 392003.58 463790.329608 457284.546328 77.046 1193969.12 1447683.33 803351 857495 52653.1 36249.3 52143.7 36294.6 32166.96 22026 180 40553 97 41620 97 12713 313 26923.25 1898482 456825 4.030 41454 40153 5223 4349 4944.236 6166.505 21411.310 6011.664 8345.197 16483.983 3047.383 36.693 279.135 46.598 71.933 0.14943384 0.00197913 416913.28 575097.864266 589946.819707 76.340 1341723.87 1549447.81 848574 885006 56898.5 39466.0 57310.6 39241.0 34665.69 23889 166 42571 93 43907 92 13587 293 29230.45 OpenBenchmarking.org
BlogBench Test: Read E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read Defaults No Mitigations 500K 1000K 1500K 2000K 2500K SE +/- 33888.90, N = 9 SE +/- 57701.61, N = 9 SE +/- 10337.24, N = 9 SE +/- 9296.08, N = 9 SE +/- 22223.70, N = 3 SE +/- 18200.59, N = 3 SE +/- 20585.69, N = 3 SE +/- 29090.95, N = 3 SE +/- 7509.49, N = 3 SE +/- 6830.31, N = 3 1830416 1906597 647728 644528 2038525 2129085 2215082 2238187 1839796 1898482 1. (CC) gcc options: -O2 -pthread
Sockperf Test: Throughput E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput Defaults No Mitigations 130K 260K 390K 520K 650K SE +/- 2385.14, N = 25 SE +/- 5568.73, N = 25 SE +/- 4980.38, N = 25 SE +/- 9010.49, N = 20 SE +/- 1976.69, N = 5 SE +/- 3747.04, N = 11 SE +/- 3227.63, N = 13 SE +/- 4690.41, N = 6 SE +/- 4840.40, N = 5 SE +/- 4338.87, N = 5 266654 340151 486761 619109 357391 417357 398750 405922 449645 456825 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Sockperf Test: Latency Ping Pong E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong Defaults No Mitigations 1.0998 2.1996 3.2994 4.3992 5.499 SE +/- 0.048, N = 9 SE +/- 0.041, N = 5 SE +/- 0.012, N = 5 SE +/- 0.013, N = 5 SE +/- 0.059, N = 5 SE +/- 0.033, N = 5 SE +/- 0.032, N = 5 SE +/- 0.026, N = 5 SE +/- 0.048, N = 5 SE +/- 0.041, N = 5 4.807 4.030 3.398 2.691 4.888 3.984 3.941 3.739 4.810 4.030 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
iPerf Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 1 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 1 Defaults No Mitigations 15K 30K 45K 60K 75K SE +/- 457.96, N = 3 SE +/- 492.87, N = 3 SE +/- 979.71, N = 3 SE +/- 209.37, N = 3 SE +/- 51.60, N = 3 SE +/- 45.67, N = 3 SE +/- 135.89, N = 3 SE +/- 115.65, N = 3 SE +/- 138.75, N = 3 SE +/- 648.27, N = 3 34129 35326 63586 70194 34314 35924 36719 36951 41194 41454 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
iPerf Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 32 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 32 Defaults No Mitigations 16K 32K 48K 64K 80K SE +/- 347.97, N = 11 SE +/- 364.79, N = 8 SE +/- 557.92, N = 3 SE +/- 569.85, N = 3 SE +/- 478.95, N = 3 SE +/- 53.38, N = 3 SE +/- 210.56, N = 3 SE +/- 117.22, N = 3 SE +/- 97.36, N = 3 SE +/- 111.16, N = 3 38523 36871 68520 72902 33418 32464 33326 33593 39761 40153 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
DaCapo Benchmark Java Test: Jython E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython Defaults No Mitigations 1200 2400 3600 4800 6000 SE +/- 41.12, N = 4 SE +/- 62.64, N = 4 SE +/- 44.17, N = 4 SE +/- 45.86, N = 6 SE +/- 49.06, N = 20 SE +/- 51.23, N = 5 SE +/- 42.13, N = 4 SE +/- 60.79, N = 4 SE +/- 5.93, N = 4 SE +/- 31.98, N = 4 5411 5371 4022 3928 4490 4390 4824 4841 5183 5223
DaCapo Benchmark Java Test: Tradesoap E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap Defaults No Mitigations 1500 3000 4500 6000 7500 SE +/- 26.70, N = 4 SE +/- 65.18, N = 3 SE +/- 45.80, N = 4 SE +/- 80.41, N = 4 SE +/- 60.00, N = 20 SE +/- 12.73, N = 4 SE +/- 66.56, N = 4 SE +/- 16.83, N = 4 SE +/- 31.20, N = 4 SE +/- 46.11, N = 8 5079 4919 6572 6361 6799 6746 6701 6543 4443 4349
Renaissance Test: Random Forest E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Random Forest Defaults No Mitigations 1200 2400 3600 4800 6000 SE +/- 46.78, N = 5 SE +/- 32.46, N = 5 SE +/- 52.75, N = 5 SE +/- 54.67, N = 5 SE +/- 52.01, N = 5 SE +/- 44.38, N = 25 SE +/- 36.85, N = 5 SE +/- 63.72, N = 25 SE +/- 63.53, N = 5 SE +/- 15.80, N = 5 5619.45 5695.30 4882.48 4730.42 5742.32 5703.45 5593.39 5685.46 5037.82 4944.24
Renaissance Test: Apache Spark ALS E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Apache Spark ALS Defaults No Mitigations 1500 3000 4500 6000 7500 SE +/- 47.83, N = 5 SE +/- 65.03, N = 8 SE +/- 40.48, N = 25 SE +/- 56.93, N = 5 SE +/- 63.46, N = 5 SE +/- 93.24, N = 5 SE +/- 85.73, N = 5 SE +/- 73.05, N = 8 SE +/- 68.93, N = 5 SE +/- 79.77, N = 5 6449.65 6468.94 6049.13 5893.83 7143.70 6983.43 6961.55 6965.36 6279.25 6166.51
Renaissance Test: Apache Spark PageRank E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Apache Spark PageRank Defaults No Mitigations 5K 10K 15K 20K 25K SE +/- 153.33, N = 5 SE +/- 102.15, N = 5 SE +/- 164.69, N = 5 SE +/- 185.35, N = 5 SE +/- 289.82, N = 7 SE +/- 321.17, N = 5 SE +/- 252.92, N = 5 SE +/- 165.93, N = 25 SE +/- 246.15, N = 5 SE +/- 142.95, N = 5 24481.44 24696.02 23626.66 23339.00 25666.99 25213.78 24250.42 24532.44 21806.95 21411.31
Renaissance Test: Twitter HTTP Requests E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Twitter HTTP Requests Defaults No Mitigations 2K 4K 6K 8K 10K SE +/- 65.80, N = 5 SE +/- 55.95, N = 5 SE +/- 34.89, N = 25 SE +/- 35.76, N = 5 SE +/- 86.77, N = 6 SE +/- 62.92, N = 5 SE +/- 93.34, N = 5 SE +/- 91.17, N = 5 SE +/- 51.36, N = 5 SE +/- 54.63, N = 5 5838.76 5725.52 5189.18 5027.58 7845.88 7749.77 7163.35 7193.92 6472.96 6011.66
Renaissance Test: In-Memory Database Shootout E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: In-Memory Database Shootout Defaults No Mitigations 3K 6K 9K 12K 15K SE +/- 38.21, N = 5 SE +/- 39.69, N = 5 SE +/- 55.86, N = 5 SE +/- 62.93, N = 5 SE +/- 185.95, N = 5 SE +/- 182.50, N = 5 SE +/- 83.54, N = 5 SE +/- 86.51, N = 25 SE +/- 18.32, N = 5 SE +/- 47.56, N = 5 7368.55 6856.78 5182.95 5007.48 16087.98 14649.67 14068.53 14030.64 8650.31 8345.20
Renaissance Test: Akka Unbalanced Cobwebbed Tree E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Akka Unbalanced Cobwebbed Tree Defaults No Mitigations 6K 12K 18K 24K 30K SE +/- 89.19, N = 25 SE +/- 114.06, N = 12 SE +/- 114.33, N = 8 SE +/- 95.58, N = 11 SE +/- 369.52, N = 25 SE +/- 273.31, N = 25 SE +/- 206.12, N = 5 SE +/- 278.89, N = 6 SE +/- 166.19, N = 5 SE +/- 120.23, N = 25 13778.51 13307.74 11118.97 10751.79 26470.85 25007.43 23067.16 22828.08 16488.15 16483.98
Renaissance Test: Genetic Algorithm Using Jenetics + Futures E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.10.0 Test: Genetic Algorithm Using Jenetics + Futures Defaults No Mitigations 1500 3000 4500 6000 7500 SE +/- 21.62, N = 5 SE +/- 8.29, N = 5 SE +/- 5.95, N = 5 SE +/- 4.16, N = 5 SE +/- 75.38, N = 5 SE +/- 41.77, N = 5 SE +/- 83.22, N = 5 SE +/- 49.14, N = 5 SE +/- 25.93, N = 5 SE +/- 24.48, N = 5 4769.20 4674.33 1677.19 1645.89 6840.66 6553.62 6607.00 6535.20 3056.52 3047.38
Timed Linux Kernel Compilation Time To Compile E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.4 Time To Compile Defaults No Mitigations 40 80 120 160 200 SE +/- 0.81, N = 3 SE +/- 0.80, N = 3 SE +/- 0.63, N = 3 SE +/- 1.18, N = 3 SE +/- 0.32, N = 13 SE +/- 0.37, N = 10 SE +/- 0.37, N = 12 SE +/- 0.35, N = 10 SE +/- 0.48, N = 5 SE +/- 0.42, N = 6 107.83 105.27 176.99 172.55 41.19 40.04 37.48 37.45 37.66 36.69
Timed LLVM Compilation Time To Compile E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 10.0 Time To Compile Defaults No Mitigations 300 600 900 1200 1500 SE +/- 4.46, N = 3 SE +/- 6.25, N = 3 SE +/- 1.45, N = 3 SE +/- 0.89, N = 3 SE +/- 4.03, N = 3 SE +/- 4.04, N = 3 SE +/- 2.40, N = 3 SE +/- 3.45, N = 3 SE +/- 3.03, N = 3 SE +/- 2.43, N = 3 812.20 809.02 1367.41 1348.51 304.45 300.31 284.99 283.60 288.26 279.14
Timed PHP Compilation Time To Compile E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.4.2 Time To Compile Defaults No Mitigations 20 40 60 80 100 SE +/- 0.19, N = 3 SE +/- 0.20, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.28, N = 3 SE +/- 0.17, N = 3 SE +/- 0.18, N = 3 SE +/- 0.10, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 82.49 80.83 99.77 98.02 50.96 50.22 48.74 48.31 46.63 46.60
Build2 Time To Compile E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.12 Time To Compile Defaults No Mitigations 50 100 150 200 250 SE +/- 0.56, N = 3 SE +/- 0.21, N = 3 SE +/- 0.33, N = 3 SE +/- 0.50, N = 3 SE +/- 0.21, N = 3 SE +/- 0.23, N = 3 SE +/- 0.30, N = 3 SE +/- 0.21, N = 3 SE +/- 0.06, N = 3 SE +/- 0.15, N = 3 142.67 140.59 223.76 219.18 80.04 78.66 75.46 75.71 72.70 71.93
Perl Benchmarks Test: Pod2html E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html Defaults No Mitigations 0.0349 0.0698 0.1047 0.1396 0.1745 SE +/- 0.00070236, N = 3 SE +/- 0.00037082, N = 3 SE +/- 0.00011071, N = 3 SE +/- 0.00025928, N = 3 SE +/- 0.00040353, N = 3 SE +/- 0.00004149, N = 3 SE +/- 0.00033629, N = 3 SE +/- 0.00038123, N = 3 SE +/- 0.00155250, N = 3 SE +/- 0.00103450, N = 3 0.15528423 0.14966118 0.11922478 0.11721965 0.12653994 0.12412203 0.11832707 0.11850206 0.14704507 0.14943384
Perl Benchmarks Test: Interpreter E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter Defaults No Mitigations 0.0008 0.0016 0.0024 0.0032 0.004 SE +/- 0.00002291, N = 3 SE +/- 0.00000822, N = 3 SE +/- 0.00000140, N = 3 SE +/- 0.00000418, N = 3 SE +/- 0.00000081, N = 3 SE +/- 0.00000968, N = 3 SE +/- 0.00000821, N = 3 SE +/- 0.00000338, N = 3 SE +/- 0.00000179, N = 3 SE +/- 0.00000873, N = 3 0.00284140 0.00247743 0.00099263 0.00093284 0.00375265 0.00330160 0.00329017 0.00325939 0.00203798 0.00197913
KeyDB E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Ops/sec, More Is Better KeyDB 5.3.1 Defaults No Mitigations 120K 240K 360K 480K 600K SE +/- 552.92, N = 3 SE +/- 2190.71, N = 3 SE +/- 780.39, N = 3 SE +/- 6242.43, N = 3 SE +/- 2599.40, N = 3 SE +/- 2174.50, N = 3 SE +/- 6677.70, N = 3 SE +/- 6409.22, N = 3 SE +/- 1324.55, N = 3 SE +/- 998.25, N = 3 448533.08 561921.38 332575.74 460535.14 365949.97 407888.25 416533.95 418381.24 392003.58 416913.28 1. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Only E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only Defaults No Mitigations 130K 260K 390K 520K 650K SE +/- 156.88, N = 3 SE +/- 228.14, N = 3 SE +/- 81.36, N = 3 SE +/- 90.34, N = 3 SE +/- 3500.01, N = 3 SE +/- 1667.57, N = 3 SE +/- 2138.58, N = 3 SE +/- 1072.68, N = 3 SE +/- 1933.16, N = 3 SE +/- 5524.79, N = 3 162112.13 176460.21 106967.31 121451.02 529354.42 564233.10 603798.29 596866.11 463790.33 575097.86 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only Defaults No Mitigations 130K 260K 390K 520K 650K SE +/- 2143.18, N = 3 SE +/- 720.15, N = 3 SE +/- 109.52, N = 3 SE +/- 73.52, N = 3 SE +/- 4246.39, N = 3 SE +/- 7109.36, N = 3 SE +/- 2731.37, N = 3 SE +/- 2508.79, N = 3 SE +/- 4489.83, N = 3 SE +/- 1562.55, N = 3 165170.20 178879.32 110061.94 121036.73 562889.82 598979.96 626099.89 621603.19 457284.55 589946.82 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
SQLite Speedtest Timed Time - Size 1,000 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 Defaults No Mitigations 20 40 60 80 100 SE +/- 0.28, N = 3 SE +/- 0.19, N = 3 SE +/- 0.07, N = 3 SE +/- 0.32, N = 3 SE +/- 0.21, N = 3 SE +/- 0.14, N = 3 SE +/- 0.05, N = 3 SE +/- 0.88, N = 3 SE +/- 0.08, N = 3 SE +/- 0.14, N = 3 88.93 73.62 68.14 57.37 77.43 66.46 68.26 68.25 77.05 76.34 1. (CC) gcc options: -O2 -ldl -lz -lpthread
Redis Test: LPUSH E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: LPUSH Defaults No Mitigations 400K 800K 1200K 1600K 2000K SE +/- 27018.39, N = 15 SE +/- 20648.66, N = 15 SE +/- 11769.75, N = 3 SE +/- 12385.25, N = 3 SE +/- 19819.76, N = 15 SE +/- 17420.09, N = 15 SE +/- 1086.39, N = 3 SE +/- 14124.93, N = 15 SE +/- 25308.80, N = 15 SE +/- 5320.14, N = 3 1280493.84 1355759.00 1709562.63 1791215.29 1321227.17 1414137.39 1371743.83 1422514.42 1193969.12 1341723.87 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Redis Test: SET E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SET Defaults No Mitigations 500K 1000K 1500K 2000K 2500K SE +/- 21392.75, N = 3 SE +/- 25094.74, N = 14 SE +/- 8659.59, N = 3 SE +/- 12548.75, N = 3 SE +/- 34935.38, N = 12 SE +/- 12486.65, N = 3 SE +/- 25451.15, N = 15 SE +/- 22461.09, N = 15 SE +/- 30115.39, N = 15 SE +/- 26730.97, N = 15 1550985.79 1605077.06 2036733.29 2175635.00 1567680.21 1607048.83 1706887.67 1665880.83 1447683.33 1549447.81 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
Facebook RocksDB Test: Random Fill E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Defaults No Mitigations 200K 400K 600K 800K 1000K SE +/- 5040.53, N = 3 SE +/- 5104.84, N = 3 SE +/- 3718.84, N = 3 SE +/- 3440.56, N = 3 SE +/- 354.47, N = 3 SE +/- 450.36, N = 3 SE +/- 171.10, N = 3 SE +/- 564.78, N = 3 SE +/- 7650.33, N = 3 SE +/- 8207.72, N = 3 518865 678788 774300 885609 141751 176835 179667 182743 803351 848574 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Facebook RocksDB Test: Sequential Fill E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill Defaults No Mitigations 200K 400K 600K 800K 1000K SE +/- 6623.34, N = 3 SE +/- 1715.59, N = 3 SE +/- 10398.91, N = 3 SE +/- 1082.27, N = 3 SE +/- 125.91, N = 3 SE +/- 239.82, N = 3 SE +/- 95.29, N = 3 SE +/- 549.70, N = 3 SE +/- 5621.96, N = 3 SE +/- 446.26, N = 3 694894 1038858 1003248 1113455 142824 177505 180768 184392 857495 885006 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Memcached mcperf Method: Get - Connections: 16 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.6.0 Method: Get - Connections: 16 Defaults No Mitigations 40K 80K 120K 160K 200K SE +/- 708.71, N = 3 SE +/- 264.58, N = 3 SE +/- 872.88, N = 3 SE +/- 2615.25, N = 3 SE +/- 3188.13, N = 11 SE +/- 326.35, N = 3 SE +/- 4357.54, N = 12 SE +/- 335.20, N = 3 SE +/- 31.75, N = 3 SE +/- 252.89, N = 3 88133.1 114148.3 122674.1 163813.9 88890.2 108534.1 101109.8 117411.7 52653.1 56898.5 1. (CC) gcc options: -O2 -lm -rdynamic
Memcached mcperf Method: Append - Connections: 16 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.6.0 Method: Append - Connections: 16 Defaults No Mitigations 20K 40K 60K 80K 100K SE +/- 300.69, N = 3 SE +/- 1113.05, N = 9 SE +/- 158.54, N = 3 SE +/- 480.34, N = 3 SE +/- 89.56, N = 3 SE +/- 84.40, N = 3 SE +/- 1545.55, N = 9 SE +/- 4883.48, N = 9 SE +/- 340.55, N = 3 SE +/- 320.11, N = 3 37763.5 49633.8 75114.2 100971.3 56358.7 67756.4 30394.7 61641.5 36249.3 39466.0 1. (CC) gcc options: -O2 -lm -rdynamic
Memcached mcperf Method: Delete - Connections: 16 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.6.0 Method: Delete - Connections: 16 Defaults No Mitigations 30K 60K 90K 120K 150K SE +/- 237.15, N = 3 SE +/- 553.89, N = 3 SE +/- 541.89, N = 3 SE +/- 212.40, N = 3 SE +/- 206.80, N = 3 SE +/- 186.49, N = 3 SE +/- 361.82, N = 3 SE +/- 352.84, N = 3 SE +/- 255.88, N = 3 SE +/- 514.16, N = 3 84443.9 112817.5 119621.9 163124.0 53759.9 57494.3 107712.7 116085.7 52143.7 57310.6 1. (CC) gcc options: -O2 -lm -rdynamic
Memcached mcperf Method: Replace - Connections: 16 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.6.0 Method: Replace - Connections: 16 Defaults No Mitigations 20K 40K 60K 80K 100K SE +/- 451.68, N = 3 SE +/- 3060.01, N = 6 SE +/- 239.83, N = 3 SE +/- 300.08, N = 3 SE +/- 1793.31, N = 9 SE +/- 73.67, N = 3 SE +/- 80.26, N = 3 SE +/- 2639.05, N = 9 SE +/- 456.00, N = 3 SE +/- 491.76, N = 5 36941.2 50251.4 75108.9 101249.7 36701.9 68552.6 29106.3 68576.5 36294.6 39241.0 1. (CC) gcc options: -O2 -lm -rdynamic
NGINX Benchmark Static Web Page Serving E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving Defaults No Mitigations 10K 20K 30K 40K 50K SE +/- 90.82, N = 3 SE +/- 286.46, N = 3 SE +/- 125.50, N = 3 SE +/- 125.28, N = 3 SE +/- 265.88, N = 3 SE +/- 34.48, N = 3 SE +/- 128.75, N = 3 SE +/- 438.85, N = 3 SE +/- 455.17, N = 4 SE +/- 57.51, N = 3 23945.41 31796.96 37413.38 46860.12 27430.11 34006.44 31476.33 34256.90 32166.96 34665.69 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
Apache HBase Test: Increment - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Rows Per Second, More Is Better Apache HBase 2.2.3 Test: Increment - Clients: 4 Defaults No Mitigations 6K 12K 18K 24K 30K SE +/- 48.15, N = 3 SE +/- 167.68, N = 3 SE +/- 312.72, N = 4 SE +/- 275.14, N = 3 SE +/- 140.39, N = 3 SE +/- 105.08, N = 3 SE +/- 183.36, N = 3 SE +/- 146.72, N = 3 SE +/- 157.49, N = 3 SE +/- 241.60, N = 3 8811 10138 23474 26591 10347 12134 12004 12059 22026 23889
Apache HBase Test: Increment - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Microseconds - Average Latency, Fewer Is Better Apache HBase 2.2.3 Test: Increment - Clients: 4 Defaults No Mitigations 100 200 300 400 500 SE +/- 2.31, N = 3 SE +/- 6.56, N = 3 SE +/- 2.39, N = 4 SE +/- 1.53, N = 3 SE +/- 5.21, N = 3 SE +/- 2.85, N = 3 SE +/- 4.84, N = 3 SE +/- 4.18, N = 3 SE +/- 1.33, N = 3 SE +/- 1.76, N = 3 452 393 169 149 384 328 332 330 180 166
Apache HBase Test: Random Read - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Rows Per Second, More Is Better Apache HBase 2.2.3 Test: Random Read - Clients: 4 Defaults No Mitigations 10K 20K 30K 40K 50K SE +/- 134.28, N = 3 SE +/- 126.04, N = 3 SE +/- 385.72, N = 10 SE +/- 431.33, N = 11 SE +/- 158.37, N = 11 SE +/- 257.81, N = 3 SE +/- 352.22, N = 3 SE +/- 90.89, N = 3 SE +/- 395.68, N = 9 SE +/- 246.18, N = 3 18668 21439 41639 47925 18054 21613 21155 21377 40553 42571
Apache HBase Test: Random Read - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Microseconds - Average Latency, Fewer Is Better Apache HBase 2.2.3 Test: Random Read - Clients: 4 Defaults No Mitigations 50 100 150 200 250 SE +/- 1.53, N = 3 SE +/- 1.15, N = 3 SE +/- 0.91, N = 10 SE +/- 0.82, N = 11 SE +/- 1.99, N = 11 SE +/- 2.33, N = 3 SE +/- 2.85, N = 3 SE +/- 0.88, N = 3 SE +/- 0.99, N = 9 212 185 95 82 219 183 187 184 97 93
Apache HBase Test: Sequential Read - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Rows Per Second, More Is Better Apache HBase 2.2.3 Test: Sequential Read - Clients: 4 Defaults No Mitigations 9K 18K 27K 36K 45K SE +/- 90.67, N = 3 SE +/- 196.14, N = 3 SE +/- 236.42, N = 13 SE +/- 646.09, N = 15 SE +/- 195.48, N = 3 SE +/- 169.00, N = 3 SE +/- 91.09, N = 3 SE +/- 293.61, N = 4 SE +/- 549.82, N = 3 SE +/- 635.18, N = 4 19519 22439 29375 36949 17630 22683 21557 22365 41620 43907
Apache HBase Test: Sequential Read - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Microseconds - Average Latency, Fewer Is Better Apache HBase 2.2.3 Test: Sequential Read - Clients: 4 Defaults No Mitigations 50 100 150 200 250 SE +/- 1.00, N = 3 SE +/- 1.53, N = 3 SE +/- 1.18, N = 13 SE +/- 2.21, N = 15 SE +/- 2.19, N = 3 SE +/- 1.15, N = 3 SE +/- 2.08, N = 4 SE +/- 1.35, N = 4 207 180 135 108 227 178 187 180 97 92
Apache HBase Test: Async Random Write - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Rows Per Second, More Is Better Apache HBase 2.2.3 Test: Async Random Write - Clients: 4 Defaults No Mitigations 4K 8K 12K 16K 20K SE +/- 37.03, N = 3 SE +/- 64.13, N = 3 SE +/- 194.46, N = 3 SE +/- 445.81, N = 12 SE +/- 5.69, N = 3 SE +/- 60.29, N = 3 SE +/- 96.81, N = 3 SE +/- 26.91, N = 3 SE +/- 135.63, N = 3 SE +/- 210.38, N = 3 5736 6651 14693 16665 6111 7938 7480 7738 12713 13587
Apache HBase Test: Async Random Write - Clients: 4 E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Microseconds - Average Latency, Fewer Is Better Apache HBase 2.2.3 Test: Async Random Write - Clients: 4 Defaults No Mitigations 150 300 450 600 750 SE +/- 4.48, N = 3 SE +/- 5.84, N = 3 SE +/- 3.61, N = 3 SE +/- 8.79, N = 12 SE +/- 0.33, N = 3 SE +/- 3.71, N = 3 SE +/- 6.94, N = 3 SE +/- 3.53, N = 3 SE +/- 4.58, N = 3 695 600 271 241 652 502 533 515 313 293
Apache Benchmark Static Web Page Serving E5-2687W v3 E3-1275 v6 2 x Gold 6138 2 x Gold 5220R EPYC 7502P OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving Defaults No Mitigations 8K 16K 24K 32K 40K SE +/- 164.03, N = 15 SE +/- 313.41, N = 4 SE +/- 64.23, N = 3 SE +/- 232.87, N = 3 SE +/- 339.27, N = 3 SE +/- 327.02, N = 3 SE +/- 395.32, N = 3 SE +/- 335.47, N = 3 SE +/- 44.61, N = 3 SE +/- 137.09, N = 3 16731.18 22745.66 27301.81 35270.52 20351.02 24738.87 24204.88 24785.26 26923.25 29230.45 1. (CC) gcc options: -shared -fPIC -O2 -pthread
Phoronix Test Suite v10.8.4