TFLite Xeon

2 x Intel Xeon Platinum 8280 testing with a GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2008240-NE-TFLITEXEO64&grt&sor.

TFLite XeonProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen Resolution4210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v32 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)378GB1920x1080OpenBenchmarking.orgProcessor Details- Scaling Governor: intel_pstate performance - CPU Microcode: 0x500002cSecurity Details- 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

TFLite Xeontensorflow-lite: SqueezeNettensorflow-lite: Inception V4tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quanttensorflow-lite: Inception ResNet V24210R v14210R v28280 v18280 v28280 v38280 2P v18280 2P v28280 2P v31547162319593155227109921113742205358015458223146401544761100521135662041053105743155493712268474575.176579.21390553102800154110311950774240.676592.01387610102847153829011855973976.275997.2138613767472.41069320114202.362934.764430.193554476156.0100599115343472332.669941.697326089144.4105583315390269538.967326.2980449OpenBenchmarking.org

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNet8280 2P v18280 2P v28280 2P v38280 v28280 v38280 v14210R v24210R v130K60K90K120K150KSE +/- 1098.16, N = 3SE +/- 1762.12, N = 15SE +/- 1941.92, N = 15SE +/- 233.29, N = 3SE +/- 126.24, N = 3SE +/- 1283.06, N = 6SE +/- 150.87, N = 3SE +/- 152.07, N = 367472.476156.089144.4102800.0102847.0105743.0154582.0154716.0

TensorFlow Lite

Model: Inception V4

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V48280 2P v28280 2P v38280 2P v18280 v38280 v28280 v14210R v24210R v1500K1000K1500K2000K2500KSE +/- 10450.58, N = 15SE +/- 15517.10, N = 3SE +/- 20326.92, N = 15SE +/- 400.67, N = 3SE +/- 774.97, N = 3SE +/- 2242.09, N = 3SE +/- 2959.04, N = 3SE +/- 1516.29, N = 310059911055833106932015382901541103155493723146402319593

TensorFlow Lite

Model: NASNet Mobile

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet Mobile8280 2P v18280 v38280 v28280 v18280 2P v28280 2P v34210R v24210R v130K60K90K120K150KSE +/- 2961.47, N = 15SE +/- 175.36, N = 3SE +/- 277.66, N = 3SE +/- 663.59, N = 3SE +/- 3072.44, N = 15SE +/- 443.76, N = 3SE +/- 283.20, N = 3SE +/- 142.99, N = 3114202.3118559.0119507.0122684.0153434.0153902.0154476.0155227.0

TensorFlow Lite

Model: Mobilenet Float

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Float8280 2P v18280 2P v38280 2P v28280 v38280 v28280 v14210R v14210R v220K40K60K80K100KSE +/- 1916.53, N = 15SE +/- 1490.28, N = 15SE +/- 647.40, N = 3SE +/- 220.54, N = 3SE +/- 161.34, N = 3SE +/- 162.15, N = 3SE +/- 66.48, N = 3SE +/- 63.93, N = 362934.769538.972332.673976.274240.674575.1109921.0110052.0

TensorFlow Lite

Model: Mobilenet Quant

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet Quant8280 2P v18280 2P v38280 2P v28280 v38280 v18280 v24210R v24210R v120K40K60K80K100KSE +/- 1621.78, N = 15SE +/- 1711.56, N = 12SE +/- 1177.33, N = 15SE +/- 133.45, N = 3SE +/- 47.48, N = 3SE +/- 98.01, N = 3SE +/- 171.12, N = 3SE +/- 97.74, N = 364430.167326.269941.675997.276579.276592.0113566.0113742.0

TensorFlow Lite

Model: Inception ResNet V2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V28280 2P v18280 2P v28280 2P v38280 v38280 v28280 v14210R v24210R v1400K800K1200K1600K2000KSE +/- 5901.31, N = 3SE +/- 8738.98, N = 3SE +/- 15046.75, N = 3SE +/- 383.68, N = 3SE +/- 788.48, N = 3SE +/- 1031.28, N = 3SE +/- 4314.23, N = 3SE +/- 9019.28, N = 393554497326098044913861371387610139055320410532053580

Phoronix Test Suite v10.8.4