Server Mitigation Costs AMD EPYC vs. Intel Xeon

Tests for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1905200-HV-XEONEPYC661&rdt&gru.

ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution2 x EPYC 76012 x Xeon 82802 x Xeon 6138 mitigations=off mitigations=auto mitigations=auto mitigations=off mitigations=auto mitigations=off mitigations=auto,nosmt2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads)Dell 02MJ3T (1.2.5 BIOS)AMD 17h516096MB120GB SSDSCKJB120G7R + 20 x 500GB Samsung SSD 860llvmpipe 504GBVE2282 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA + 2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 19.045.0.0-15-generic (x86_64)GNOME Shell 3.32.0X Server 1.20.4modesetting 1.20.43.3 Mesa 19.0.2 (LLVM 8.0 128 bits)GCC 8.3.0ext41600x120020 x 500GB Samsung SSD 860 + 120GB SSDSCKJB120G7R2 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)Intel Sky Lake-E DMI3 Registers386048MBSamsung SSD 970 PRO 512GBllvmpipe 377GB2 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE3.3 Mesa 19.0.2 (LLVM 8.0 256 bits)1920x10802 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads)TYAN S7106 (V1.02.B20 BIOS)96256MBSamsung SSD 970 EVO 250GBASPEED2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE + Mellanox MT26448llvmpipe 93GB3.3 Mesa 19.0.2 (LLVM 8.0 256 bits)2 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores)OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v Disk Details- 2 x EPYC 7601: mitigations=off: MQ-DEADLINE / errors=remount-ro,relatime,rw- 2 x EPYC 7601: mitigations=auto: MQ-DEADLINE / errors=remount-ro,relatime,rw- 2 x Xeon 8280: mitigations=auto: NONE / errors=remount-ro,relatime,rw- 2 x Xeon 8280: mitigations=off: NONE / errors=remount-ro,relatime,rw- 2 x Xeon 6138: mitigations=auto: NONE / errors=remount-ro,relatime,rw- 2 x Xeon 6138: mitigations=off: NONE / errors=remount-ro,relatime,rw- 2 x Xeon 6138: mitigations=auto,nosmt: NONE / errors=remount-ro,relatime,rwJava Details- 2 x EPYC 7601: mitigations=off, 2 x EPYC 7601: mitigations=auto: OpenJDK Runtime Environment (build 11.0.3+7-Ubuntu-1ubuntu219.04.1)Python Details- Python 2.7.16 + Python 3.7.3Security Details- 2 x EPYC 7601: mitigations=off: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled- 2 x EPYC 7601: mitigations=auto: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling- 2 x Xeon 8280: mitigations=auto: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling- 2 x Xeon 8280: mitigations=off: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled- 2 x Xeon 6138: mitigations=auto: KPTI + 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 __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling - 2 x Xeon 6138: mitigations=off: l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled- 2 x Xeon 6138: mitigations=auto,nosmt: KPTI + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT disabled + mds: Mitigation of Clear buffers; SMT disabled + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW RSB filling Processor Details- 2 x Xeon 8280: mitigations=auto, 2 x Xeon 8280: mitigations=off, 2 x Xeon 6138: mitigations=auto, 2 x Xeon 6138: mitigations=off, 2 x Xeon 6138: mitigations=auto,nosmt: Scaling Governor: intel_pstate powersave

stress-ng: Semaphoresstress-ng: Socket Activitystress-ng: Context Switchingstress-ng: System V Message Passingethr: TCP - Connections/s - 1sysbench: Memorysysbench: CPUfs-mark: 1000 Files, 1MB Sizeblogbench: Readblogbench: Writex264: H.264 Video Encodingv-ray: CPUindigobench: Bedroomindigobench: Supercarcompilebench: Compilecompilebench: Initial Createethr: HTTP - Bandwidth - 1ethr: TCP - Bandwidth - 64ethr: UDP - Bandwidth - 64ethr: HTTP - Bandwidth - 64tjbench: Decompression Throughputsockperf: Throughputcompress-7zip: Compress Speed Testmcperf: Addmcperf: Getmcperf: Setmcperf: Appendmcperf: Deletemcperf: Prependmcperf: Replaceredis: LPOPredis: SADDredis: LPUSHredis: GETredis: SETnginx: Static Web Page Servingapache: Static Web Page Servingphpbench: PHP Benchmark Suiteopenssl: RSA 4096-bit Performancepostmark: Disk Transaction Performancepgbench: Buffer Test - Normal Load - Read Onlyapache-siege: 250ctx-clock: Context Switch Timenamd: ATPase Simulation - 327,506 Atomsethr: TCP - Latency - 1ethr: TCP - Latency - 32ethr: TCP - Latency - 64pybench: Total For Average Test Timesglibc-bench: ffsglibc-bench: ffsllglibc-bench: pthread_onceosbench: Memory Allocationst-test1: 2build-linux-kernel: Time To Compilebuild-llvm: Time To Compiley-cruncher: Calculating 500M Pi Digitscompress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19dav1d: Summer Nature 4Kdav1d: Summer Nature 1080phackbench: 32 - Processperl-benchmark: Interpreterdarktable: Boat - CPU-onlydarktable: Masskrug - CPU-onlydarktable: Server Room - CPU-onlygimp: resizegimp: rotategimp: auto-levelsgimp: unsharp-maskgit: Time To Complete Common Git Commandsscikit-learn: osbench: Create Filesosbench: Create Threadsosbench: Launch Programssockperf: Latency Ping Pongsockperf: Latency Under Load2 x EPYC 76012 x Xeon 82802 x Xeon 6138 mitigations=off mitigations=auto mitigations=auto mitigations=off mitigations=auto mitigations=off mitigations=auto,nosmt87990129655295211969053163915741785588922717278610227144140409123.245.42159839467935028611313210771393850501255554146564508406063449149533346543460420001501647554113639417605941335784183901712439313192245434517604226201320.4482720.0219.2318.9214712.382.382.3892.749.3623.2913515.1127.0414.4619.006.5542.200.0012428010.956.824.7012.8618.2131.9540.217.1012.5118.3465.1093.153.1114.6188104730310193901688747607998341955178871916880574327972138413243.315.34147837666635446011338010731393388871326803972259792397704267861095424234664617221431589297112427017230671302734199471580739364092375247516768223791320.4474221.4322.8920.4714852.422.422.4291.648.9523.6313215.2725.9214.3219.366.4743.410.0012550610.746.885.2713.0518.0130.8239.897.1512.6919.5864.1091.663.6213.4821345483362928606228620362765101199030994825160119662696.149.881823568422405980280802145717346079523920363698107699639657213010819573459298772651345211750316446942425725183082030480254046414541360659529102491310.3580219.5018.2518.8910171.661.661.6785.398.3321.221059.1321.539.0417.819.1736.090.003155073.904.531.9012.3918.3022.5027.045.9027.7013.7119.5853.883.7337.2821384634038962103866582285366871299184196578179116663326.119.86186262944340017228292814511734746252420116641411922667582676721203077910177823234278221068201688954239669818528273227628637641228137076199916749344471300.3581818.7716.6917.1010161.651.651.6685.308.8921.191059.1021.819.1318.019.0536.450.003123373.984.541.9812.1418.0722.7325.705.8327.4712.8919.2254.213.5434.2718356371915021838073505211255871256619253630164210813326553109391473.607.821616503407291438279950173816138988714165851481907725191159608925465962160050224599918248321515264223782516237922845021705595173784448075436085770.6612421.8721.5721.9610792.062.062.0698.969.1231.7615311.5523.2410.8425.0511.9252.030.003572715.184.882.2113.3519.5626.9132.116.3729.7916.5122.5956.644.6028.5318926042351733171910547926268211232563854116174213669925399109394903.637.9117605944602991712726071856160477593143127625371167716144374928117500749777509116985482124682169005422858841718403354542683359592378665859589830315291010.6642116.1316.9716.0710821.771.781.7794.268.9830.4515211.4823.0610.7024.8011.8543.930.003124655.004.532.1912.7918.8025.1829.676.2029.7613.5820.1552.853.7632.44190416911463747466777910185350710770637362190222039927105132314412.626.49161950240925353327975220001604000951191244705792579494066037495114609826074921669632109222165885423226441846430285172234459701175454901432205296825860.7316620.4520.6420.4710842.062.072.0698.1222.5534.8916610.9323.2910.7425.6613.1849.610.003526844.565.122.1713.4819.7025.8730.856.2929.7416.4119.9449.274.6538.58OpenBenchmarking.org

Stress-NG

Test: Semaphores

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Semaphoresmitigations=offmitigations=automitigations=auto,nosmt500K1000K1500K2000K2500KSE +/- 800.35, N = 3SE +/- 1591.51, N = 3SE +/- 34922.09, N = 3SE +/- 31145.04, N = 4SE +/- 11950.24, N = 3SE +/- 26899.91, N = 3SE +/- 14400.89, N = 3879901881047213454821384631835637189260419041691. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Socket Activity

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Socket Activitymitigations=offmitigations=automitigations=auto,nosmt9K18K27K36K45KSE +/- 73.80, N = 3SE +/- 404.44, N = 3SE +/- 379.29, N = 15SE +/- 96.68, N = 3SE +/- 164.54, N = 15SE +/- 315.22, N = 3SE +/- 101.22, N = 11296553031033629403891915023517114631. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Context Switching

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Context Switchingmitigations=offmitigations=automitigations=auto,nosmt13M26M39M52M65MSE +/- 569562.26, N = 15SE +/- 352165.21, N = 15SE +/- 1845916.70, N = 13SE +/- 2456469.23, N = 12SE +/- 612779.97, N = 15SE +/- 414106.69, N = 3SE +/- 69516.88, N = 1029521196193901682860622862103866218380733317191074746671. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: System V Message Passing

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: System V Message Passingmitigations=offmitigations=automitigations=auto,nosmt2M4M6M8M10MSE +/- 58001.23, N = 3SE +/- 139629.62, N = 3SE +/- 45437.62, N = 3SE +/- 41897.18, N = 3SE +/- 71564.53, N = 13SE +/- 90049.91, N = 15SE +/- 45489.89, N = 390531638747607620362758228535052112547926277910181. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Ethr

Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1mitigations=offmitigations=automitigations=auto,nosmt2K4K6K8K10KSE +/- 750.57, N = 12SE +/- 143.61, N = 4SE +/- 87.18, N = 5SE +/- 64.38, N = 3SE +/- 6.67, N = 3SE +/- 59.94, N = 15SE +/- 65.06, N = 39157998365106687558768215350

Sysbench

Test: Memory

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 2018-07-28Test: Memorymitigations=offmitigations=automitigations=auto,nosmt3M6M9M12M15MSE +/- 14365.40, N = 3SE +/- 29225.92, N = 3SE +/- 204934.51, N = 3SE +/- 155544.19, N = 3SE +/- 84700.36, N = 3SE +/- 164167.54, N = 5SE +/- 35095.29, N = 3417855841955171199030912991841125661921232563871077061. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=core2 -rdynamic -ldl -laio -lm

Sysbench

Test: CPU

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 2018-07-28Test: CPUmitigations=offmitigations=automitigations=auto,nosmt20K40K60K80K100KSE +/- 321.16, N = 3SE +/- 185.02, N = 3SE +/- 246.13, N = 3SE +/- 234.65, N = 3SE +/- 299.85, N = 3SE +/- 287.90, N = 3SE +/- 149.78, N = 3892278871994825965785363054116373621. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=core2 -rdynamic -ldl -laio -lm

FS-Mark

Test: 1000 Files, 1MB Size

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 1000 Files, 1MB Sizemitigations=offmitigations=automitigations=auto,nosmt4080120160200SE +/- 0.10, N = 3SE +/- 2.74, N = 3SE +/- 1.32, N = 3SE +/- 0.60, N = 3SE +/- 1.30, N = 3SE +/- 2.01, N = 12SE +/- 0.99, N = 31721681601791641741901. (CC) gcc options: -static

BlogBench

Test: Read

2 x EPYC 76012 x Xeon 6138OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: Readmitigations=offmitigations=automitigations=auto,nosmt500K1000K1500K2000K2500KSE +/- 4332.00, N = 3SE +/- 11227.28, N = 4SE +/- 25869.91, N = 4SE +/- 25190.15, N = 6SE +/- 17169.51, N = 37861028057432108133213669922203991. (CC) gcc options: -O2 -pthread

BlogBench

Test: Write

2 x EPYC 76012 x Xeon 6138OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: Writemitigations=offmitigations=automitigations=auto,nosmt6K12K18K24K30KSE +/- 579.82, N = 3SE +/- 876.12, N = 3SE +/- 1128.88, N = 3SE +/- 1394.85, N = 3SE +/- 1108.45, N = 327144279722655325399271051. (CC) gcc options: -O2 -pthread

x264

H.264 Video Encoding

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2018-09-25H.264 Video Encodingmitigations=offmitigations=automitigations=auto,nosmt306090120150SE +/- 4.68, N = 12SE +/- 4.45, N = 12SE +/- 0.98, N = 15SE +/- 0.72, N = 3SE +/- 0.87, N = 15SE +/- 1.81, N = 3SE +/- 1.22, N = 101401381191161091091321. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize

Chaos Group V-RAY

Mode: CPU

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgKsamples, More Is BetterChaos Group V-RAY 4.10.03Mode: CPUmitigations=offmitigations=automitigations=auto,nosmt14K28K42K56K70KSE +/- 168.86, N = 3SE +/- 276.21, N = 3SE +/- 287.66, N = 3SE +/- 367.13, N = 3SE +/- 14.11, N = 3SE +/- 88.98, N = 3SE +/- 345.97, N = 740912413246626966332391473949031441

IndigoBench

Scene: Bedroom

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.0.64Scene: Bedroommitigations=offmitigations=automitigations=auto,nosmt246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 33.243.316.146.113.603.632.62

IndigoBench

Scene: Supercar

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.0.64Scene: Supercarmitigations=offmitigations=automitigations=auto,nosmt3691215SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 35.425.349.889.867.827.916.49

Compile Bench

Test: Compile

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Compilemitigations=offmitigations=automitigations=auto,nosmt400800120016002000SE +/- 2.12, N = 3SE +/- 14.96, N = 3SE +/- 15.24, N = 3SE +/- 29.31, N = 3SE +/- 10.79, N = 3SE +/- 16.97, N = 3SE +/- 10.24, N = 31598147818231862161617601619

Compile Bench

Test: Initial Create

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial Createmitigations=offmitigations=automitigations=auto,nosmt140280420560700SE +/- 1.43, N = 3SE +/- 3.79, N = 3SE +/- 29.78, N = 3SE +/- 12.93, N = 3SE +/- 3.65, N = 3SE +/- 1.38, N = 3SE +/- 6.14, N = 3394376568629503594502

Ethr

Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1mitigations=offmitigations=automitigations=auto,nosmt150300450600750SE +/- 5.43, N = 13SE +/- 8.76, N = 12SE +/- 4.50, N = 3SE +/- 3.90, N = 11SE +/- 2.69, N = 3SE +/- 5.31, N = 3SE +/- 1.64, N = 3679666422443407460409

Ethr

Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64mitigations=offmitigations=automitigations=auto,nosmt90K180K270K360K450KSE +/- 5512.12, N = 15SE +/- 5315.55, N = 15SE +/- 3252.90, N = 15SE +/- 5495.09, N = 4SE +/- 3729.07, N = 15SE +/- 3356.98, N = 7SE +/- 2244.19, N = 15350286354460405980400172291438299171253533

Ethr

Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64mitigations=offmitigations=automitigations=auto,nosmt60K120K180K240K300KSE +/- 1206.05, N = 3SE +/- 621.31, N = 3SE +/- 2773.77, N = 15SE +/- 4051.11, N = 4SE +/- 2845.78, N = 3SE +/- 838.93, N = 3SE +/- 3957.23, N = 4113132113380280802282928279950272607279752

Ethr

Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 64

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 64mitigations=offmitigations=automitigations=auto,nosmt400800120016002000SE +/- 22.31, N = 13SE +/- 21.35, N = 15SE +/- 17.07, N = 3SE +/- 10.88, N = 3SE +/- 8.98, N = 3SE +/- 22.55, N = 3SE +/- 17.97, N = 31077107314571451173818562000

libjpeg-turbo tjbench

Test: Decompression Throughput

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 1.5.3Test: Decompression Throughputmitigations=offmitigations=automitigations=auto,nosmt4080120160200SE +/- 0.02, N = 3SE +/- 0.16, N = 3SE +/- 0.62, N = 3SE +/- 0.79, N = 3SE +/- 0.18, N = 3SE +/- 0.83, N = 3SE +/- 0.61, N = 31391391731731611601601. (CC) gcc options: -O3 -lm

Sockperf

Test: Throughput

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMessages Per Second, More Is BetterSockperf 3.4Test: Throughputmitigations=offmitigations=automitigations=auto,nosmt100K200K300K400K500KSE +/- 2279.86, N = 5SE +/- 3843.01, N = 5SE +/- 3697.13, N = 13SE +/- 5646.61, N = 5SE +/- 3112.70, N = 5SE +/- 2094.96, N = 5SE +/- 2813.72, N = 53850503388874607954746253898874775934000951. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread

7-Zip Compression

Compress Speed Test

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed Testmitigations=offmitigations=automitigations=auto,nosmt50K100K150K200K250KSE +/- 1861.30, N = 15SE +/- 1629.96, N = 15SE +/- 378.50, N = 3SE +/- 2140.92, N = 3SE +/- 1921.18, N = 3SE +/- 1846.23, N = 3SE +/- 717.49, N = 31255551326802392032420111416581431271191241. (CXX) g++ options: -pipe -lpthread

Memcached mcperf

Method: Add

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Addmitigations=offmitigations=automitigations=auto,nosmt14K28K42K56K70KSE +/- 405.74, N = 9SE +/- 293.10, N = 3SE +/- 158.50, N = 3SE +/- 37.41, N = 3SE +/- 314.75, N = 3SE +/- 468.12, N = 3SE +/- 288.90, N = 3414653972263698664145148162537470571. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Get

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Getmitigations=offmitigations=automitigations=auto,nosmt30K60K90K120K150KSE +/- 887.55, N = 15SE +/- 2153.62, N = 12SE +/- 1143.77, N = 3SE +/- 1539.50, N = 3SE +/- 1878.69, N = 15SE +/- 595.59, N = 3SE +/- 1747.94, N = 15645085979210769911922690772116771925791. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Set

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Setmitigations=offmitigations=automitigations=auto,nosmt14K28K42K56K70KSE +/- 532.26, N = 3SE +/- 152.02, N = 3SE +/- 120.94, N = 3SE +/- 639.02, N = 3SE +/- 810.72, N = 3SE +/- 757.70, N = 3SE +/- 784.89, N = 3406063977063965675825191161443494061. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Append

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Appendmitigations=offmitigations=automitigations=auto,nosmt16K32K48K64K80KSE +/- 264.91, N = 3SE +/- 77.68, N = 3SE +/- 262.96, N = 3SE +/- 3815.61, N = 15SE +/- 237.85, N = 3SE +/- 232.30, N = 3SE +/- 645.72, N = 3344914267872130676725960874928603741. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Delete

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Deletemitigations=offmitigations=automitigations=auto,nosmt30K60K90K120K150KSE +/- 477.00, N = 3SE +/- 573.05, N = 3SE +/- 171.11, N = 3SE +/- 340.72, N = 3SE +/- 574.38, N = 3SE +/- 466.08, N = 3SE +/- 889.77, N = 3495336109510819512030792546117500951141. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Prepend

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Prependmitigations=offmitigations=automitigations=auto,nosmt20K40K60K80K100KSE +/- 173.68, N = 3SE +/- 261.61, N = 3SE +/- 284.89, N = 3SE +/- 124.09, N = 3SE +/- 171.47, N = 3SE +/- 274.42, N = 3SE +/- 659.82, N = 3346544242373459791015962174977609821. (CC) gcc options: -O2 -lm -rdynamic

Memcached mcperf

Method: Replace

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgOperations Per Second, More Is BetterMemcached mcperf 1.5.10Method: Replacemitigations=offmitigations=automitigations=auto,nosmt20K40K60K80K100KSE +/- 357.59, N = 3SE +/- 2561.52, N = 15SE +/- 463.79, N = 3SE +/- 199.42, N = 3SE +/- 166.90, N = 3SE +/- 578.68, N = 3SE +/- 179.68, N = 3346044664629877778236005075091607491. (CC) gcc options: -O2 -lm -rdynamic

Redis

Test: LPOP

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: LPOPmitigations=offmitigations=automitigations=auto,nosmt600K1200K1800K2400K3000KSE +/- 12284.48, N = 3SE +/- 102668.04, N = 12SE +/- 35393.09, N = 15SE +/- 113061.70, N = 12SE +/- 52651.60, N = 12SE +/- 22295.59, N = 15SE +/- 118587.03, N = 1220001501722143265134523427822245999169854821669631. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: SADD

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: SADDmitigations=offmitigations=automitigations=auto,nosmt500K1000K1500K2000K2500KSE +/- 9402.91, N = 3SE +/- 15869.82, N = 3SE +/- 27846.46, N = 4SE +/- 28497.18, N = 15SE +/- 35772.75, N = 15SE +/- 29843.36, N = 15SE +/- 32224.84, N = 316475541589297211750321068201824832212468221092221. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: LPUSH

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: LPUSHmitigations=offmitigations=automitigations=auto,nosmt400K800K1200K1600K2000KSE +/- 4158.06, N = 3SE +/- 12003.04, N = 3SE +/- 26605.02, N = 3SE +/- 17264.97, N = 15SE +/- 27411.17, N = 15SE +/- 22011.67, N = 4SE +/- 20018.38, N = 311363941124270164469416889541515264169005416588541. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: GET

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: GETmitigations=offmitigations=automitigations=auto,nosmt500K1000K1500K2000K2500KSE +/- 19423.29, N = 6SE +/- 28093.95, N = 3SE +/- 24817.10, N = 3SE +/- 32913.22, N = 15SE +/- 58120.66, N = 13SE +/- 51401.87, N = 12SE +/- 27581.42, N = 317605941723067242572523966982237825228588423226441. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: SET

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: SETmitigations=offmitigations=automitigations=auto,nosmt400K800K1200K1600K2000KSE +/- 17082.01, N = 4SE +/- 7460.25, N = 3SE +/- 21788.12, N = 15SE +/- 33328.41, N = 15SE +/- 30299.72, N = 15SE +/- 21579.08, N = 15SE +/- 15900.51, N = 313357841302734183082018528271623792171840318464301. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

NGINX Benchmark

Static Web Page Serving

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterNGINX Benchmark 1.9.9Static Web Page Servingmitigations=offmitigations=automitigations=auto,nosmt8K16K24K32K40KSE +/- 99.43, N = 3SE +/- 452.04, N = 15SE +/- 141.49, N = 3SE +/- 119.83, N = 3SE +/- 120.63, N = 3SE +/- 219.72, N = 3SE +/- 56.90, N = 3183901994730480322762845035454285171. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native

Apache Benchmark

Static Web Page Serving

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page Servingmitigations=offmitigations=automitigations=auto,nosmt6K12K18K24K30KSE +/- 122.50, N = 3SE +/- 271.36, N = 12SE +/- 245.97, N = 9SE +/- 218.04, N = 3SE +/- 150.57, N = 3SE +/- 399.53, N = 3SE +/- 73.97, N = 3171241580725404286372170526833223441. (CC) gcc options: -shared -fPIC -O2 -pthread

PHPBench

PHP Benchmark Suite

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suitemitigations=offmitigations=automitigations=auto,nosmt140K280K420K560K700KSE +/- 206.79, N = 3SE +/- 1026.75, N = 3SE +/- 1036.09, N = 3SE +/- 1035.82, N = 3SE +/- 1594.73, N = 3SE +/- 842.63, N = 3SE +/- 763.78, N = 3393131393640641454641228595173595923597011

OpenSSL

RSA 4096-bit Performance

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit Performancemitigations=offmitigations=automitigations=auto,nosmt3K6K9K12K15KSE +/- 7.33, N = 3SE +/- 10.48, N = 3SE +/- 105.64, N = 3SE +/- 13.37, N = 3SE +/- 64.00, N = 3SE +/- 87.77, N = 3SE +/- 82.48, N = 39224923713606137077844786675451. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

PostMark

Disk Transaction Performance

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction Performancemitigations=offmitigations=automitigations=auto,nosmt13002600390052006500SE +/- 72.33, N = 3SE +/- 51.00, N = 3SE +/- 46.33, N = 354345247595261994807585949011. (CC) gcc options: -O3

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Normal Load - Mode: Read Onlymitigations=offmitigations=automitigations=auto,nosmt200K400K600K800K1000KSE +/- 7486.91, N = 3SE +/- 2651.13, N = 3SE +/- 13866.36, N = 3SE +/- 5069.09, N = 3SE +/- 1103.67, N = 3SE +/- 4024.98, N = 3SE +/- 295.06, N = 35176045167689102499167495436085898304322051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Apache Siege

Concurrent Users: 250

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 250mitigations=offmitigations=automitigations=auto,nosmt7K14K21K28K35KSE +/- 167.66, N = 3SE +/- 150.02, N = 3SE +/- 114.29, N = 3SE +/- 518.60, N = 3SE +/- 121.61, N = 322620223793444731529296821. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto

ctx_clock

Context Switch Time

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch Timemitigations=offmitigations=automitigations=auto,nosmt130260390520650SE +/- 0.67, N = 3SE +/- 0.67, N = 3SE +/- 0.67, N = 3SE +/- 1.15, N = 3132132131130577101586

NAMD

ATPase Simulation - 327,506 Atoms

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13b1ATPase Simulation - 327,506 Atomsmitigations=offmitigations=automitigations=auto,nosmt0.16460.32920.49380.65840.823SE +/- 0.00013, N = 3SE +/- 0.00036, N = 3SE +/- 0.00004, N = 3SE +/- 0.00023, N = 3SE +/- 0.00045, N = 3SE +/- 0.00214, N = 3SE +/- 0.00072, N = 30.448270.447420.358020.358180.661240.664210.73166

Ethr

Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1mitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.80, N = 15SE +/- 1.04, N = 12SE +/- 0.59, N = 12SE +/- 0.98, N = 15SE +/- 0.40, N = 15SE +/- 0.16, N = 3SE +/- 0.25, N = 1520.0221.4319.5018.7721.8716.1320.45

Ethr

Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 32

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 32mitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.46, N = 15SE +/- 1.06, N = 12SE +/- 0.28, N = 3SE +/- 0.29, N = 15SE +/- 0.30, N = 15SE +/- 0.33, N = 15SE +/- 0.27, N = 1519.2322.8918.2516.6921.5716.9720.64

Ethr

Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64mitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.35, N = 12SE +/- 0.68, N = 15SE +/- 0.46, N = 15SE +/- 0.45, N = 15SE +/- 0.32, N = 15SE +/- 0.27, N = 3SE +/- 0.24, N = 1518.9220.4718.8917.1021.9616.0720.47

PyBench

Total For Average Test Times

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test Timesmitigations=offmitigations=automitigations=auto,nosmt30060090012001500SE +/- 4.70, N = 3SE +/- 17.37, N = 3SE +/- 2.40, N = 3SE +/- 1.53, N = 3SE +/- 1.45, N = 3SE +/- 3.00, N = 3SE +/- 1.15, N = 31471148510171016107910821084

glibc bench

Benchmark: ffs

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: ffsmitigations=offmitigations=automitigations=auto,nosmt0.54451.0891.63352.1782.7225SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.382.421.661.652.061.772.06

glibc bench

Benchmark: ffsll

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: ffsllmitigations=offmitigations=automitigations=auto,nosmt0.54451.0891.63352.1782.7225SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.382.421.661.652.061.782.07

glibc bench

Benchmark: pthread_once

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: pthread_oncemitigations=offmitigations=automitigations=auto,nosmt0.54451.0891.63352.1782.7225SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.382.421.671.662.061.772.06

OSBench

Test: Memory Allocations

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory Allocationsmitigations=offmitigations=automitigations=auto,nosmt20406080100SE +/- 0.20, N = 3SE +/- 0.16, N = 3SE +/- 0.45, N = 3SE +/- 0.22, N = 3SE +/- 0.19, N = 3SE +/- 0.13, N = 3SE +/- 0.03, N = 392.7491.6485.3985.3098.9694.2698.121. (CC) gcc options: -lm

t-test1

Threads: 2

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 2mitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 39.368.958.338.899.128.9822.551. (CC) gcc options: -pthread

Timed Linux Kernel Compilation

Time To Compile

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.18Time To Compilemitigations=offmitigations=automitigations=auto,nosmt816243240SE +/- 0.23, N = 13SE +/- 0.19, N = 15SE +/- 0.33, N = 14SE +/- 0.33, N = 14SE +/- 0.41, N = 13SE +/- 0.30, N = 15SE +/- 0.36, N = 1423.2923.6321.2221.1931.7630.4534.89

Timed LLVM Compilation

Time To Compile

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 6.0.1Time To Compilemitigations=offmitigations=automitigations=auto,nosmt4080120160200135132105105153152166

Y-Cruncher

Calculating 500M Pi Digits

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.7Calculating 500M Pi Digitsmitigations=offmitigations=automitigations=auto,nosmt48121620SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 315.1115.279.139.1011.5511.4810.93

XZ Compression

Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterXZ Compression 5.2.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9mitigations=offmitigations=automitigations=auto,nosmt612182430SE +/- 0.61, N = 15SE +/- 0.24, N = 15SE +/- 0.31, N = 4SE +/- 0.24, N = 15SE +/- 0.32, N = 3SE +/- 0.09, N = 3SE +/- 0.26, N = 327.0425.9221.5321.8123.2423.0623.291. (CC) gcc options: -pthread -fvisibility=hidden -O2

Zstd Compression

Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterZstd Compression 1.3.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19mitigations=offmitigations=automitigations=auto,nosmt48121620SE +/- 0.27, N = 15SE +/- 0.39, N = 15SE +/- 0.11, N = 4SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 15SE +/- 0.02, N = 314.4614.329.049.1310.8410.7010.741. (CC) gcc options: -O3 -pthread -lz

dav1d

Video Input: Summer Nature 4K

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is Betterdav1d 0.3Video Input: Summer Nature 4Kmitigations=offmitigations=automitigations=auto,nosmt612182430SE +/- 0.08, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 319.0019.3617.8118.0125.0524.8025.661. (CC) gcc options: -pthread

dav1d

Video Input: Summer Nature 1080p

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is Betterdav1d 0.3Video Input: Summer Nature 1080pmitigations=offmitigations=automitigations=auto,nosmt3691215SE +/- 0.08, N = 3SE +/- 0.08, N = 4SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 36.556.479.179.0511.9211.8513.181. (CC) gcc options: -pthread

Hackbench

Count: 32 - Type: Process

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: Processmitigations=offmitigations=automitigations=auto,nosmt1224364860SE +/- 0.46, N = 15SE +/- 0.55, N = 3SE +/- 0.40, N = 15SE +/- 0.58, N = 3SE +/- 1.34, N = 15SE +/- 0.41, N = 3SE +/- 0.43, N = 1542.2043.4136.0936.4552.0343.9349.611. (CC) gcc options: -lpthread

Perl Benchmarks

Test: Interpreter

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterPerl BenchmarksTest: Interpretermitigations=offmitigations=automitigations=auto,nosmt0.00080.00160.00240.00320.004SE +/- 0.00001750, N = 3SE +/- 0.00001075, N = 3SE +/- 0.00000704, N = 3SE +/- 0.00001217, N = 3SE +/- 0.00000502, N = 3SE +/- 0.00000617, N = 3SE +/- 0.00001529, N = 30.001242800.001255060.003155070.003123370.003572710.003124650.00352684

Darktable

Test: Boat - Acceleration: CPU-only

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.6.0Test: Boat - Acceleration: CPU-onlymitigations=offmitigations=automitigations=auto,nosmt3691215SE +/- 0.64, N = 12SE +/- 0.81, N = 12SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 5SE +/- 0.03, N = 3SE +/- 0.07, N = 1510.9510.743.903.985.185.004.56

Darktable

Test: Masskrug - Acceleration: CPU-only

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.6.0Test: Masskrug - Acceleration: CPU-onlymitigations=offmitigations=automitigations=auto,nosmt246810SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 15SE +/- 0.10, N = 15SE +/- 0.03, N = 36.826.884.534.544.884.535.12

Darktable

Test: Server Room - Acceleration: CPU-only

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 2.6.0Test: Server Room - Acceleration: CPU-onlymitigations=offmitigations=automitigations=auto,nosmt1.18582.37163.55744.74325.929SE +/- 0.30, N = 15SE +/- 0.02, N = 3SE +/- 0.02, N = 15SE +/- 0.03, N = 15SE +/- 0.03, N = 5SE +/- 0.00, N = 3SE +/- 0.03, N = 154.705.271.901.982.212.192.17

GIMP

Test: resize

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.8Test: resizemitigations=offmitigations=automitigations=auto,nosmt3691215SE +/- 0.13, N = 15SE +/- 0.21, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.13, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 312.8613.0512.3912.1413.3512.7913.48

GIMP

Test: rotate

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.8Test: rotatemitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.20, N = 818.2118.0118.3018.0719.5618.8019.70

GIMP

Test: auto-levels

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.8Test: auto-levelsmitigations=offmitigations=automitigations=auto,nosmt714212835SE +/- 0.20, N = 3SE +/- 0.26, N = 15SE +/- 0.24, N = 7SE +/- 0.33, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.10, N = 331.9530.8222.5022.7326.9125.1825.87

GIMP

Test: unsharp-mask

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.8Test: unsharp-maskmitigations=offmitigations=automitigations=auto,nosmt918273645SE +/- 0.28, N = 3SE +/- 0.22, N = 3SE +/- 0.25, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.52, N = 340.2139.8927.0425.7032.1129.6730.85

Git

Time To Complete Common Git Commands

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterGitTime To Complete Common Git Commandsmitigations=offmitigations=automitigations=auto,nosmt246810SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 37.107.155.905.836.376.206.291. git version 2.20.1

Scikit-Learn

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 0.17.1mitigations=offmitigations=automitigations=auto,nosmt714212835SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.13, N = 3SE +/- 0.07, N = 312.5112.6927.7027.4729.7929.7629.74

OSBench

Test: Create Files

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create Filesmitigations=offmitigations=automitigations=auto,nosmt510152025SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 318.3419.5813.7112.8916.5113.5816.411. (CC) gcc options: -lm

OSBench

Test: Create Threads

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create Threadsmitigations=offmitigations=automitigations=auto,nosmt1530456075SE +/- 0.77, N = 5SE +/- 0.50, N = 3SE +/- 0.13, N = 3SE +/- 0.15, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 365.1064.1019.5819.2222.5920.1519.941. (CC) gcc options: -lm

OSBench

Test: Launch Programs

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch Programsmitigations=offmitigations=automitigations=auto,nosmt20406080100SE +/- 0.51, N = 3SE +/- 0.66, N = 3SE +/- 1.28, N = 12SE +/- 0.41, N = 3SE +/- 0.97, N = 3SE +/- 0.39, N = 3SE +/- 0.70, N = 393.1591.6653.8854.2156.6452.8549.271. (CC) gcc options: -lm

Sockperf

Test: Latency Ping Pong

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.4Test: Latency Ping Pongmitigations=offmitigations=automitigations=auto,nosmt1.04632.09263.13894.18525.2315SE +/- 0.02, N = 5SE +/- 0.04, N = 5SE +/- 0.01, N = 5SE +/- 0.02, N = 5SE +/- 0.04, N = 25SE +/- 0.03, N = 11SE +/- 0.04, N = 253.113.623.733.544.603.764.651. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread

Sockperf

Test: Latency Under Load

2 x EPYC 76012 x Xeon 82802 x Xeon 6138OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.4Test: Latency Under Loadmitigations=offmitigations=automitigations=auto,nosmt918273645SE +/- 0.55, N = 25SE +/- 0.42, N = 25SE +/- 0.55, N = 25SE +/- 1.78, N = 20SE +/- 2.22, N = 25SE +/- 1.33, N = 20SE +/- 0.36, N = 2514.6113.4837.2834.2728.5332.4438.581. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread

Phoronix Test Suite v10.8.5