TFLite Xeon

2 x AMD EPYC 7282 16-Core testing with a Supermicro H11DSi-NT v2.00 (2.1 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2009067-NE-TFLITEXEO66
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
4210R v1
August 23 2020
  24 Minutes
4210R v2
August 23 2020
  24 Minutes
8280 v1
August 23 2020
  23 Minutes
8280 v2
August 23 2020
  20 Minutes
8280 v3
August 23 2020
  20 Minutes
8280 2P v1
August 24 2020
  1 Hour, 8 Minutes
8280 2P v2
August 24 2020
  1 Hour, 7 Minutes
8280 2P v3
August 24 2020
  52 Minutes
7282 2P
September 06 2020
  31 Minutes
Invert Hiding All Results Option
  37 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


TFLite XeonProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen Resolution4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P2 x Intel Xeon Silver 4210R @ 3.20GHz (20 Cores / 40 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)Intel Sky Lake-E DMI3 Registers378GB280GB INTEL SSDPE21D280GAASPEEDVE2282 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbEUbuntu 20.045.8.0-050800rc6daily20200721-generic (x86_64) 20200720GNOME Shell 3.36.1X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext41920x1080Intel Xeon Platinum 8280 @ 4.00GHz (28 Cores / 56 Threads)188GB1024x7682 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)378GB1920x10802 x AMD EPYC 7282 16-Core @ 2.80GHz (32 Cores / 64 Threads)Supermicro H11DSi-NT v2.00 (2.1 BIOS)AMD Starship/Matisse474GBVE2282 x Intel 10G X550TOpenBenchmarking.orgProcessor Details- 4210R v1: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 4210R v2: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 v1: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 v2: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 v3: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 2P v1: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 2P v2: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 8280 2P v3: Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002c- 7282 2P: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301034Security Details- 4210R v1: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 4210R v2: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 v1: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 v2: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 v3: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 2P v1: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 2P v2: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 8280 2P v3: 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 + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 7282 2P: 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: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2PResult OverviewPhoronix Test Suite100%133%165%198%231%TensorFlow LiteTensorFlow LiteTensorFlow LiteTensorFlow LiteTensorFlow LiteTensorFlow LiteInception V4SqueezeNetMobilenet FloatMobilenet QuantI.R.VNASNet Mobile

TFLite Xeontensorflow-lite: Inception V4tensorflow-lite: SqueezeNettensorflow-lite: Mobilenet Floattensorflow-lite: Inception ResNet V2tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Quant4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P2319593154716109921205358015522711374223146401545821100522041053154476113566155493710574374575.1139055312268476579.2154110310280074240.6138761011950776592.0153829010284773976.2138613711855975997.2106932067472.462934.7935544114202.364430.1100599176156.072332.697326015343469941.6105583389144.469538.998044915390267326.2105415380937.248021.993634510879949668.3OpenBenchmarking.org

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V44210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P500K1000K1500K2000K2500KSE +/- 1516.29, N = 3SE +/- 2959.04, N = 3SE +/- 2242.09, N = 3SE +/- 774.97, N = 3SE +/- 400.67, N = 3SE +/- 20326.92, N = 15SE +/- 10450.58, N = 15SE +/- 15517.10, N = 3SE +/- 1284.35, N = 3231959323146401554937154110315382901069320100599110558331054153
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V44210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P400K800K1200K1600K2000KMin: 2316570 / Avg: 2319593.33 / Max: 2321310Min: 2310760 / Avg: 2314640 / Max: 2320450Min: 1552430 / Avg: 1554936.67 / Max: 1559410Min: 1539590 / Avg: 1541103.33 / Max: 1542150Min: 1537490 / Avg: 1538290 / Max: 1538730Min: 953728 / Avg: 1069320.13 / Max: 1197720Min: 965195 / Avg: 1005990.53 / Max: 1114170Min: 1027970 / Avg: 1055833.33 / Max: 1081600Min: 1052660 / Avg: 1054153.33 / Max: 1056710

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNet4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P30K60K90K120K150KSE +/- 152.07, N = 3SE +/- 150.87, N = 3SE +/- 1283.06, N = 6SE +/- 233.29, N = 3SE +/- 126.24, N = 3SE +/- 1098.16, N = 3SE +/- 1762.12, N = 15SE +/- 1941.92, N = 15SE +/- 1025.33, N = 15154716.0154582.0105743.0102800.0102847.067472.476156.089144.480937.2
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNet4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P30K60K90K120K150KMin: 154429 / Avg: 154715.67 / Max: 154947Min: 154282 / Avg: 154582 / Max: 154760Min: 104097 / Avg: 105742.83 / Max: 112140Min: 102400 / Avg: 102799.67 / Max: 103208Min: 102683 / Avg: 102846.67 / Max: 103095Min: 65731.6 / Avg: 67472.43 / Max: 69502.6Min: 67290.2 / Avg: 76155.95 / Max: 95618.5Min: 73953.5 / Avg: 89144.35 / Max: 96745.8Min: 78724.6 / Avg: 80937.21 / Max: 92728.6

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Float4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P20K40K60K80K100KSE +/- 66.48, N = 3SE +/- 63.93, N = 3SE +/- 162.15, N = 3SE +/- 161.34, N = 3SE +/- 220.54, N = 3SE +/- 1916.53, N = 15SE +/- 647.40, N = 3SE +/- 1490.28, N = 15SE +/- 60.48, N = 3109921.0110052.074575.174240.673976.262934.772332.669538.948021.9
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Float4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P20K40K60K80K100KMin: 109816 / Avg: 109920.67 / Max: 110044Min: 109935 / Avg: 110052.33 / Max: 110155Min: 74250.8 / Avg: 74575.1 / Max: 74737.5Min: 73920.6 / Avg: 74240.6 / Max: 74436.6Min: 73545 / Avg: 73976.23 / Max: 74272.1Min: 49562.5 / Avg: 62934.66 / Max: 75437.7Min: 71038 / Avg: 72332.6 / Max: 72999.7Min: 58519.9 / Avg: 69538.91 / Max: 77146.3Min: 47922 / Avg: 48021.87 / Max: 48130.9

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V24210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P400K800K1200K1600K2000KSE +/- 9019.28, N = 3SE +/- 4314.23, N = 3SE +/- 1031.28, N = 3SE +/- 788.48, N = 3SE +/- 383.68, N = 3SE +/- 5901.31, N = 3SE +/- 8738.98, N = 3SE +/- 15046.75, N = 3SE +/- 990.14, N = 320535802041053139055313876101386137935544973260980449936345
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V24210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P400K800K1200K1600K2000KMin: 2040410 / Avg: 2053580 / Max: 2070840Min: 2034200 / Avg: 2041053.33 / Max: 2049020Min: 1389390 / Avg: 1390553.33 / Max: 1392610Min: 1386220 / Avg: 1387610 / Max: 1388950Min: 1385600 / Avg: 1386136.67 / Max: 1386880Min: 924598 / Avg: 935544 / Max: 944840Min: 955926 / Avg: 973260.33 / Max: 983864Min: 953272 / Avg: 980449 / Max: 1005230Min: 934494 / Avg: 936345 / Max: 937880

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet Mobile4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P30K60K90K120K150KSE +/- 142.99, N = 3SE +/- 283.20, N = 3SE +/- 663.59, N = 3SE +/- 277.66, N = 3SE +/- 175.36, N = 3SE +/- 2961.47, N = 15SE +/- 3072.44, N = 15SE +/- 443.76, N = 3SE +/- 389.84, N = 3155227.0154476.0122684.0119507.0118559.0114202.3153434.0153902.0108799.0
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet Mobile4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P30K60K90K120K150KMin: 155054 / Avg: 155227.33 / Max: 155511Min: 153973 / Avg: 154476 / Max: 154953Min: 121357 / Avg: 122683.67 / Max: 123379Min: 118973 / Avg: 119507 / Max: 119906Min: 118379 / Avg: 118559.33 / Max: 118910Min: 95589.9 / Avg: 114202.33 / Max: 133504Min: 135284 / Avg: 153433.6 / Max: 176482Min: 153383 / Avg: 153902 / Max: 154785Min: 108087 / Avg: 108799.33 / Max: 109430

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Quant4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P20K40K60K80K100KSE +/- 97.74, N = 3SE +/- 171.12, N = 3SE +/- 47.48, N = 3SE +/- 98.01, N = 3SE +/- 133.45, N = 3SE +/- 1621.78, N = 15SE +/- 1177.33, N = 15SE +/- 1711.56, N = 12SE +/- 81.92, N = 3113742.0113566.076579.276592.075997.264430.169941.667326.249668.3
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Quant4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v37282 2P20K40K60K80K100KMin: 113567 / Avg: 113741.67 / Max: 113905Min: 113231 / Avg: 113565.67 / Max: 113795Min: 76486 / Avg: 76579.2 / Max: 76641.6Min: 76401.5 / Avg: 76591.97 / Max: 76727.3Min: 75752.6 / Avg: 75997.17 / Max: 76212Min: 53845.3 / Avg: 64430.14 / Max: 75867.7Min: 62904.8 / Avg: 69941.63 / Max: 76355.8Min: 58964.5 / Avg: 67326.23 / Max: 78575.1Min: 49576.1 / Avg: 49668.3 / Max: 49831.7