sdfa AMD Ryzen Threadripper 3990X 64-Core testing with a Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS) and AMD Radeon RX 5700 8GB on Ubuntu 23.10 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 2312122-PTS-SDFA911983 a Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aJava Notes: OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10)Python Notes: Python 3.11.6Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
b c d Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 23.10, Kernel: 6.5.0-13-generic (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.2.1-1ubuntu3 (LLVM 15.0.7 DRM 3.54), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160
sdfa OpenBenchmarking.org Phoronix Test Suite AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads) Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS) AMD Starship/Matisse 128GB Samsung SSD 970 EVO Plus 500GB AMD Radeon RX 5700 8GB (1750/875MHz) AMD Navi 10 HDMI Audio DELL P2415Q Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 23.10 6.5.0-13-generic (x86_64) GNOME Shell 45.0 X Server + Wayland 4.6 Mesa 23.2.1-1ubuntu3 (LLVM 15.0.7 DRM 3.54) GCC 13.2.0 ext4 3840x2160 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution Sdfa Benchmarks System Logs - Transparent Huge Pages: madvise - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107a - OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10) - Python 3.11.6 - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
a b c d Result Overview Phoronix Test Suite 100% 108% 116% 125% 133% Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Apache Spark TPC-H Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H Apache Spark TPC-H R.5.S.I - S.S.S R.5.S.I - S.S.S C.S.9.P.Y.P - S.S.S C.S.9.P.Y.P - S.S.S B.L.N.Q.A - S.S.S B.L.N.Q.A - S.S.S 10 - G.M.O.A.Q N.D.C.o.b.u.o.I - A.M.S N.D.C.o.b.u.o.I - A.M.S N.T.C.D.m - S.S.S N.T.C.D.m - S.S.S C.D.Y.C - A.M.S C.D.Y.C - A.M.S N.D.C.o.b.u.o.I - S.S.S N.D.C.o.b.u.o.I - S.S.S R.5.S.I - A.M.S R.5.S.I - A.M.S N.T.C.B.b.u.S.S.I - S.S.S N.T.C.B.b.u.S.S.I - S.S.S R.5.B - S.S.S R.5.B - S.S.S 1 - Q01 1 - Q02 1 - Q03 1 - Q04 1 - Q05 1 - Q06 1 - Q07 1 - Q08 1 - Q09 1 - Q10 1 - Q11 1 - Q12 1 - Q13 1 - Q14 1 - Q15 1 - Q16 1 - Q17 1 - Q18 1 - Q19 1 - Q20 1 - Q21 1 - Q22 10 - Q04 10 - Q05 10 - Q06 10 - Q08 10 - Q09 10 - Q11 10 - Q13 10 - Q15 10 - Q16 10 - Q18 10 - Q19 10 - Q22 50 - Q03 50 - Q05 50 - Q11 50 - Q13 50 - Q16 50 - Q22
sdfa spark-tpch: 1 - Geometric Mean Of All Queries spark-tpch: 1 - Q01 spark-tpch: 1 - Q02 spark-tpch: 1 - Q03 spark-tpch: 1 - Q04 spark-tpch: 1 - Q05 spark-tpch: 1 - Q06 spark-tpch: 1 - Q07 spark-tpch: 1 - Q08 spark-tpch: 1 - Q09 spark-tpch: 1 - Q10 spark-tpch: 1 - Q11 spark-tpch: 1 - Q12 spark-tpch: 1 - Q13 spark-tpch: 1 - Q14 spark-tpch: 1 - Q15 spark-tpch: 1 - Q16 spark-tpch: 1 - Q17 spark-tpch: 1 - Q18 spark-tpch: 1 - Q19 spark-tpch: 1 - Q20 spark-tpch: 1 - Q21 spark-tpch: 1 - Q22 spark-tpch: 10 - Geometric Mean Of All Queries spark-tpch: 10 - Q01 spark-tpch: 10 - Q02 spark-tpch: 10 - Q03 spark-tpch: 10 - Q04 spark-tpch: 10 - Q05 spark-tpch: 10 - Q06 spark-tpch: 10 - Q07 spark-tpch: 10 - Q08 spark-tpch: 10 - Q09 spark-tpch: 10 - Q10 spark-tpch: 10 - Q11 spark-tpch: 10 - Q12 spark-tpch: 10 - Q13 spark-tpch: 10 - Q14 spark-tpch: 10 - Q15 spark-tpch: 10 - Q16 spark-tpch: 10 - Q17 spark-tpch: 10 - Q18 spark-tpch: 10 - Q19 spark-tpch: 10 - Q20 spark-tpch: 10 - Q21 spark-tpch: 10 - Q22 spark-tpch: 50 - Geometric Mean Of All Queries spark-tpch: 50 - Q01 spark-tpch: 50 - Q02 spark-tpch: 50 - Q03 spark-tpch: 50 - Q04 spark-tpch: 50 - Q05 spark-tpch: 50 - Q06 spark-tpch: 50 - Q07 spark-tpch: 50 - Q08 spark-tpch: 50 - Q09 spark-tpch: 50 - Q10 spark-tpch: 50 - Q11 spark-tpch: 50 - Q12 spark-tpch: 50 - Q13 spark-tpch: 50 - Q14 spark-tpch: 50 - Q15 spark-tpch: 50 - Q16 spark-tpch: 50 - Q17 spark-tpch: 50 - Q18 spark-tpch: 50 - Q19 spark-tpch: 50 - Q20 spark-tpch: 50 - Q21 spark-tpch: 50 - Q22 deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Baseline - Synchronous Single-Stream deepsparse: ResNet-50, Baseline - Synchronous Single-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Stream deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream a b c d 2.20328734 4.69511843 2.19916511 3.15890259 2.96026391 3.46156806 0.75870752 3.16296500 2.29091269 4.40090740 2.83789939 1.30849856 2.14987269 1.40058664 1.83231536 1.99491587 1.42731291 2.66192222 4.00372249 1.02847038 2.75555176 7.39505470 1.08502284 8.17090667 8.67833742 4.85099522 10.24406719 8.63682970 12.47607358 4.40131362 10.63118744 10.99399408 14.89943695 10.26555188 4.54931625 8.03465001 4.45708307 5.79615275 5.83709431 4.32050626 12.45044740 14.86223253 7.21627538 8.84160964 29.31493378 3.53586213 27.97562823 26.02234904 10.48352242 36.61449941 30.58487765 41.16542689 20.38117282 35.63054784 38.35942205 46.89715449 35.92603938 9.06239065 30.43039259 12.33711910 25.48705292 23.04547437 10.58654118 55.60917791 62.50857417 24.71872266 30.70769501 130.86564128 10.67486382 37.9294 836.9172 13.8987 71.9375 980.4497 32.6019 129.7041 7.7066 441.4407 72.4193 103.5643 9.6462 2555.4232 12.4868 627.7539 1.5897 201.9130 158.1544 86.3769 11.5611 49.2180 649.9348 11.3088 88.4675 449.3079 71.1719 106.4432 9.3853 215.8297 147.9620 91.1432 10.9635 336.1917 95.0151 56.7902 17.5978 43.1906 736.7739 17.0745 58.5511 503.4132 63.4770 64.7072 15.4510 37.7351 841.4259 14.0751 71.0390 2.24531415 4.73092079 2.20726132 3.2391243 3.09257221 3.3776865 0.73223764 3.30348134 2.32780766 4.51510048 3.07959628 1.36447799 2.0294199 1.45866323 1.78157055 2.17460775 1.43939853 2.72264194 4.0458951 1.1739459 2.58762598 7.70369244 0.98886323 7.89399455 8.70151806 4.88495016 10.32122612 8.70809078 12.91888523 4.62162924 10.531744 10.67045593 14.1802969 10.18445492 4.26761341 8.10859299 4.03698015 5.82840872 5.68089533 4.15591764 12.44107723 14.87680531 6.0425024 8.87663841 28.79863548 3.40110779 27.78682893 25.99035454 10.45031643 36.02916718 30.26870918 43.25001144 20.29723358 35.34214783 37.97156906 46.31754303 36.27207947 8.69320774 30.40716553 12.13110542 25.39692879 22.88924599 10.53230286 55.27999115 61.9550209 25.07553101 30.53813362 131.45526123 10.52690792 39.3833 806.3928 14.4029 69.4174 993.6037 32.1567 131.167 7.6192 450.8062 70.9424 106.9295 9.3424 2635.5729 12.1095 599.1863 1.6656 206.9627 154.3035 88.165 11.3279 49.3328 648.524 11.6057 86.1505 448.7603 71.2672 106.5387 9.377 212.7552 150.2259 89.8018 11.1265 337.0821 94.8818 59.0082 16.9355 43.8384 726.2504 17.9912 55.5621 508.5902 62.8112 64.3985 15.522 37.9456 837.4581 14.3078 69.8784 2.23705911 4.84034109 2.27846766 3.34556897 2.94428595 3.52418653 0.71082556 3.18430217 2.35779921 4.32967218 3.02914357 1.28974267 2.04039538 1.39991681 1.83836599 2.14363774 1.49448760 2.58527072 3.99324067 1.20837911 2.75284823 7.46197589 1.05880324 8.11142575 8.66681448 4.92897081 10.26445103 8.50457668 12.66337935 4.30036068 10.62009271 10.85271359 14.81898117 10.41971207 4.38963429 8.17116038 4.09777006 5.86177413 5.81021436 4.17572975 12.41860390 16.11452039 6.54328155 8.91217136 29.03367678 3.69966158 28.11219413 26.00166639 10.56094869 37.05844625 30.22341601 39.70042801 20.34083811 35.45530701 37.80742772 47.10843913 36.82142385 8.75260639 30.88582865 13.13648542 25.45622126 22.94638952 10.82443651 55.33429082 62.29955419 24.89447530 30.78961944 131.42175293 10.63667170 37.7845 841.3435 13.8813 72.0273 988.4905 32.3392 132.1047 7.5689 445.1623 71.8091 106.0333 9.4215 2548.3771 12.5199 591.2570 1.6887 199.2162 160.2780 87.9306 11.3576 49.0805 649.7550 11.8387 84.4587 452.0401 70.7399 107.1042 9.3277 216.3914 147.4889 91.1565 10.9619 337.7813 94.6789 57.1793 17.4779 43.0454 737.9777 16.9865 58.8526 506.1330 63.1295 64.1154 15.5926 37.9380 837.1418 14.1048 70.8989 2.22356446 4.84622987 2.27512868 3.15447322 2.85856040 3.30995369 0.74262398 3.25330043 2.22529737 4.55872504 3.14210590 1.29190469 2.06954416 1.39225551 1.83219139 2.27715731 1.53928820 2.57773050 3.86921136 0.99435820 2.74797662 7.40773551 1.06113847 8.24567273 8.73231697 4.79437430 10.15224552 8.33958340 12.65661907 4.36710962 10.74541600 10.81727695 14.92097505 10.35872300 4.48057460 8.09419775 4.61340586 5.73522472 5.85565599 4.34601260 12.40058359 14.91129049 8.02795347 8.82452583 28.90806135 3.59682607 28.12394226 26.11903827 10.44357236 35.99200439 30.55707931 38.99460093 20.38204702 35.40038427 38.38075257 46.99718602 35.92824809 9.01768939 30.04180272 13.68457921 25.42378616 23.00236766 10.91404247 55.42181269 62.73364003 25.02892049 30.51781400 131.00612386 10.26505438 38.7188 820.1208 13.9020 71.9192 989.0589 32.3108 134.3975 7.4402 441.3941 72.4572 103.9791 9.6084 2543.1138 12.5483 620.2412 1.6090 201.9444 158.1065 87.4654 11.4172 49.1687 650.6931 11.7423 85.1572 450.0598 71.0506 104.2930 9.5802 215.2909 148.2827 89.5918 11.1539 335.0968 95.4458 56.8599 17.5761 43.0569 737.8524 17.0878 58.5011 504.8420 63.2965 64.0842 15.6001 37.4534 845.4438 13.9997 71.4165 OpenBenchmarking.org
Apache Spark TPC-H This is a benchmark of Apache Spark using TPC-H data-set. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmarks the Apache Spark in a single-system configuration using spark-submit. The test makes use of https://github.com/ssavvides/tpch-spark/ for facilitating the TPC-H benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Geometric Mean Of All Queries a b c d 0.5052 1.0104 1.5156 2.0208 2.526 SE +/- 0.02461970, N = 4 SE +/- 0.00573499, N = 3 SE +/- 0.02256519, N = 3 2.20328734 2.24531415 2.23705911 2.22356446 MIN: 0.99 / MAX: 7.56 MIN: 0.99 / MAX: 7.7 MIN: 1.05 / MAX: 7.66 MIN: 0.97 / MAX: 7.7 1Q01
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q01 a b c d 1.0904 2.1808 3.2712 4.3616 5.452 SE +/- 0.06902799, N = 4 SE +/- 0.05428360, N = 3 SE +/- 0.08250324, N = 3 4.69511843 4.73092079 4.84034109 4.84622987 1Q02
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q02 a b c d 0.5127 1.0254 1.5381 2.0508 2.5635 SE +/- 0.05229177, N = 4 SE +/- 0.08231744, N = 3 SE +/- 0.07468701, N = 3 2.19916511 2.20726132 2.27846766 2.27512868 1Q03
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q03 a b c d 0.7528 1.5056 2.2584 3.0112 3.764 SE +/- 0.03420168, N = 4 SE +/- 0.05826206, N = 3 SE +/- 0.04118145, N = 3 3.15890259 3.23912430 3.34556897 3.15447322 1Q04
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q04 a b c d 0.6958 1.3916 2.0874 2.7832 3.479 SE +/- 0.02567628, N = 4 SE +/- 0.05472939, N = 3 SE +/- 0.09994253, N = 3 2.96026391 3.09257221 2.94428595 2.85856040 1Q05
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q05 a b c d 0.7929 1.5858 2.3787 3.1716 3.9645 SE +/- 0.06297205, N = 4 SE +/- 0.08593638, N = 3 SE +/- 0.06648494, N = 3 3.46156806 3.37768650 3.52418653 3.30995369 1Q06
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q06 a b c d 0.1707 0.3414 0.5121 0.6828 0.8535 SE +/- 0.03196645, N = 4 SE +/- 0.00867146, N = 3 SE +/- 0.01813313, N = 3 0.75870752 0.73223764 0.71082556 0.74262398 1Q07
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q07 a b c d 0.7433 1.4866 2.2299 2.9732 3.7165 SE +/- 0.07379513, N = 4 SE +/- 0.13350733, N = 3 SE +/- 0.03596564, N = 3 3.16296500 3.30348134 3.18430217 3.25330043 1Q08
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q08 a b c d 0.5305 1.061 1.5915 2.122 2.6525 SE +/- 0.02917526, N = 4 SE +/- 0.06735723, N = 3 SE +/- 0.07315965, N = 3 2.29091269 2.32780766 2.35779921 2.22529737 1Q09
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q09 a b c d 1.0257 2.0514 3.0771 4.1028 5.1285 SE +/- 0.13363421, N = 4 SE +/- 0.11520840, N = 3 SE +/- 0.17668063, N = 3 4.40090740 4.51510048 4.32967218 4.55872504 1Q10
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q10 a b c d 0.707 1.414 2.121 2.828 3.535 SE +/- 0.04789727, N = 4 SE +/- 0.03625533, N = 3 SE +/- 0.04198555, N = 3 2.83789939 3.07959628 3.02914357 3.14210590 1Q11
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q11 a b c d 0.307 0.614 0.921 1.228 1.535 SE +/- 0.07929998, N = 4 SE +/- 0.15013623, N = 3 SE +/- 0.10958276, N = 3 1.30849856 1.36447799 1.28974267 1.29190469 1Q12
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q12 a b c d 0.4837 0.9674 1.4511 1.9348 2.4185 SE +/- 0.15031140, N = 4 SE +/- 0.04470734, N = 3 SE +/- 0.02411619, N = 3 2.14987269 2.02941990 2.04039538 2.06954416 1Q13
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q13 a b c d 0.3282 0.6564 0.9846 1.3128 1.641 SE +/- 0.03676867, N = 4 SE +/- 0.04981792, N = 3 SE +/- 0.02733553, N = 3 1.40058664 1.45866323 1.39991681 1.39225551 1Q14
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q14 a b c d 0.4136 0.8272 1.2408 1.6544 2.068 SE +/- 0.03540160, N = 4 SE +/- 0.01014717, N = 3 SE +/- 0.02463739, N = 3 1.83231536 1.78157055 1.83836599 1.83219139 1Q15
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q15 a b c d 0.5124 1.0248 1.5372 2.0496 2.562 SE +/- 0.04340326, N = 4 SE +/- 0.20579960, N = 3 SE +/- 0.09001964, N = 3 1.99491587 2.17460775 2.14363774 2.27715731 1Q16
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q16 a b c d 0.3463 0.6926 1.0389 1.3852 1.7315 SE +/- 0.01144007, N = 4 SE +/- 0.05940937, N = 3 SE +/- 0.04799671, N = 3 1.42731291 1.43939853 1.49448760 1.53928820 1Q17
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q17 a b c d 0.6126 1.2252 1.8378 2.4504 3.063 SE +/- 0.02035298, N = 4 SE +/- 0.01865956, N = 3 SE +/- 0.03392729, N = 3 2.66192222 2.72264194 2.58527072 2.57773050 1Q18
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q18 a b c d 0.9103 1.8206 2.7309 3.6412 4.5515 SE +/- 0.10303949, N = 4 SE +/- 0.10019808, N = 3 SE +/- 0.13038086, N = 3 4.00372249 4.04589510 3.99324067 3.86921136 1Q19
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q19 a b c d 0.2719 0.5438 0.8157 1.0876 1.3595 SE +/- 0.02636608, N = 4 SE +/- 0.07878807, N = 3 SE +/- 0.01874751, N = 3 1.02847038 1.17394590 1.20837911 0.99435820 1Q20
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q20 a b c d 0.62 1.24 1.86 2.48 3.1 SE +/- 0.09583966, N = 4 SE +/- 0.13759046, N = 3 SE +/- 0.14206756, N = 3 2.75555176 2.58762598 2.75284823 2.74797662 1Q21
1Q22
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q22 a b c d 0.2441 0.4882 0.7323 0.9764 1.2205 SE +/- 0.00972834, N = 4 SE +/- 0.00495537, N = 3 SE +/- 0.01300436, N = 3 1.08502284 0.98886323 1.05880324 1.06113847
Result
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Geometric Mean Of All Queries a b c d 2 4 6 8 10 SE +/- 0.00516651, N = 3 SE +/- 0.05385571, N = 3 SE +/- 0.01945474, N = 3 8.17090667 7.89399455 8.11142575 8.24567273 MIN: 3.42 / MAX: 29.75 MIN: 3.4 / MAX: 28.8 MIN: 3.43 / MAX: 29.17 MIN: 3.57 / MAX: 29.28 10Q01
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q01 a b c d 2 4 6 8 10 SE +/- 0.07219277, N = 3 SE +/- 0.10886176, N = 3 SE +/- 0.05844458, N = 3 8.67833742 8.70151806 8.66681448 8.73231697 10Q02
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q02 a b c d 1.109 2.218 3.327 4.436 5.545 SE +/- 0.02777661, N = 3 SE +/- 0.06191683, N = 3 SE +/- 0.02874460, N = 3 4.85099522 4.88495016 4.92897081 4.79437430 10Q03
10Q04
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q04 a b c d 2 4 6 8 10 SE +/- 0.24804165, N = 3 SE +/- 0.08517991, N = 3 SE +/- 0.07436974, N = 3 8.63682970 8.70809078 8.50457668 8.33958340 10Q05
10Q06
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q06 a b c d 1.0399 2.0798 3.1197 4.1596 5.1995 SE +/- 0.04431026, N = 3 SE +/- 0.01079337, N = 3 SE +/- 0.03080177, N = 3 4.40131362 4.62162924 4.30036068 4.36710962 10Q07
10Q08
10Q09
10Q10
10Q11
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q11 a b c d 1.0236 2.0472 3.0708 4.0944 5.118 SE +/- 0.03938644, N = 3 SE +/- 0.02158043, N = 3 SE +/- 0.00931035, N = 3 4.54931625 4.26761341 4.38963429 4.48057460 10Q12
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q12 a b c d 2 4 6 8 10 SE +/- 0.04371011, N = 3 SE +/- 0.01728126, N = 3 SE +/- 0.05148904, N = 3 8.03465001 8.10859299 8.17116038 8.09419775 10Q13
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q13 a b c d 1.038 2.076 3.114 4.152 5.19 SE +/- 0.30907989, N = 3 SE +/- 0.05603992, N = 3 SE +/- 0.25379241, N = 3 4.45708307 4.03698015 4.09777006 4.61340586 10Q14
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q14 a b c d 1.3189 2.6378 3.9567 5.2756 6.5945 SE +/- 0.02725146, N = 3 SE +/- 0.02997105, N = 3 SE +/- 0.04489420, N = 3 5.79615275 5.82840872 5.86177413 5.73522472 10Q15
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q15 a b c d 1.3175 2.635 3.9525 5.27 6.5875 SE +/- 0.03384145, N = 3 SE +/- 0.03487590, N = 3 SE +/- 0.04663640, N = 3 5.83709431 5.68089533 5.81021436 5.85565599 10Q16
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q16 a b c d 0.9779 1.9558 2.9337 3.9116 4.8895 SE +/- 0.04048184, N = 3 SE +/- 0.08541494, N = 3 SE +/- 0.00250658, N = 3 4.32050626 4.15591764 4.17572975 4.34601260 10Q17
10Q18
10Q19
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q19 a b c d 2 4 6 8 10 SE +/- 0.61100502, N = 3 SE +/- 0.63365910, N = 3 SE +/- 0.02955483, N = 3 7.21627538 6.04250240 6.54328155 8.02795347 10Q20
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q20 a b c d 2 4 6 8 10 SE +/- 0.06394497, N = 3 SE +/- 0.01360476, N = 3 SE +/- 0.08793082, N = 3 8.84160964 8.87663841 8.91217136 8.82452583 10Q21
10Q22
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q22 a b c d 0.8324 1.6648 2.4972 3.3296 4.162 SE +/- 0.10541256, N = 3 SE +/- 0.18733559, N = 3 SE +/- 0.01543789, N = 3 3.53586213 3.40110779 3.69966158 3.59682607
Scale Factor: 100
a: The test quit with a non-zero exit status. E: ./spark-tpch: line 6: echo: write error: No space left on device
b: The test quit with a non-zero exit status. E: ./spark-tpch: line 6: echo: write error: No space left on device
c: The test quit with a non-zero exit status. E: ./spark-tpch: line 6: echo: write error: No space left on device
d: The test quit with a non-zero exit status. E: ./spark-tpch: line 6: echo: write error: No space left on device
Neural Magic DeepSparse This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a b c d 9 18 27 36 45 SE +/- 0.52, N = 3 SE +/- 0.38, N = 6 SE +/- 0.36, N = 3 37.93 39.38 37.78 38.72
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a b c d 200 400 600 800 1000 SE +/- 12.88, N = 3 SE +/- 8.13, N = 6 SE +/- 7.99, N = 3 836.92 806.39 841.34 820.12
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b c d 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 13.90 14.40 13.88 13.90
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b c d 16 32 48 64 80 SE +/- 0.30, N = 3 SE +/- 0.25, N = 3 SE +/- 0.18, N = 3 71.94 69.42 72.03 71.92
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 200 400 600 800 1000 SE +/- 3.44, N = 3 SE +/- 1.16, N = 3 SE +/- 1.17, N = 3 980.45 993.60 988.49 989.06
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 8 16 24 32 40 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 32.60 32.16 32.34 32.31
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 30 60 90 120 150 SE +/- 0.69, N = 3 SE +/- 1.51, N = 4 SE +/- 1.48, N = 5 129.70 131.17 132.10 134.40
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 2 4 6 8 10 SE +/- 0.0410, N = 3 SE +/- 0.0865, N = 4 SE +/- 0.0842, N = 5 7.7066 7.6192 7.5689 7.4402
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 600 1200 1800 2400 3000 SE +/- 14.52, N = 3 SE +/- 12.01, N = 3 SE +/- 15.60, N = 3 2555.42 2635.57 2548.38 2543.11
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 140 280 420 560 700 SE +/- 2.01, N = 3 SE +/- 8.33, N = 3 SE +/- 1.41, N = 3 627.75 599.19 591.26 620.24
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 0.38 0.76 1.14 1.52 1.9 SE +/- 0.0051, N = 3 SE +/- 0.0240, N = 3 SE +/- 0.0037, N = 3 1.5897 1.6656 1.6887 1.6090
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a b c d 50 100 150 200 250 SE +/- 1.28, N = 3 SE +/- 1.60, N = 3 SE +/- 1.27, N = 3 201.91 206.96 199.22 201.94
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream a b c d 40 80 120 160 200 SE +/- 0.94, N = 3 SE +/- 1.36, N = 3 SE +/- 1.02, N = 3 158.15 154.30 160.28 158.11
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream a b c d 11 22 33 44 55 SE +/- 0.03, N = 3 SE +/- 0.26, N = 3 SE +/- 0.10, N = 3 49.22 49.33 49.08 49.17
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream a b c d 140 280 420 560 700 SE +/- 0.29, N = 3 SE +/- 1.51, N = 3 SE +/- 1.27, N = 3 649.93 648.52 649.76 650.69
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream a b c d 3 6 9 12 15 SE +/- 0.07, N = 15 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 11.31 11.61 11.84 11.74
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream a b c d 20 40 60 80 100 SE +/- 0.60, N = 15 SE +/- 0.42, N = 3 SE +/- 0.61, N = 3 88.47 86.15 84.46 85.16
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b c d 100 200 300 400 500 SE +/- 1.01, N = 3 SE +/- 0.60, N = 3 SE +/- 1.54, N = 3 449.31 448.76 452.04 450.06
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b c d 16 32 48 64 80 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 SE +/- 0.25, N = 3 71.17 71.27 70.74 71.05
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a b c d 20 40 60 80 100 SE +/- 0.16, N = 3 SE +/- 0.34, N = 3 SE +/- 1.00, N = 3 106.44 106.54 107.10 104.29
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream a b c d 3 6 9 12 15 SE +/- 0.0142, N = 3 SE +/- 0.0294, N = 3 SE +/- 0.0916, N = 3 9.3853 9.3770 9.3277 9.5802
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 50 100 150 200 250 SE +/- 0.62, N = 3 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 215.83 212.76 216.39 215.29
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 30 60 90 120 150 SE +/- 0.45, N = 3 SE +/- 0.12, N = 3 SE +/- 0.06, N = 3 147.96 150.23 147.49 148.28
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 20 40 60 80 100 SE +/- 0.36, N = 3 SE +/- 0.17, N = 3 SE +/- 0.36, N = 3 91.14 89.80 91.16 89.59
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 10.96 11.13 10.96 11.15
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a b c d 70 140 210 280 350 SE +/- 0.53, N = 3 SE +/- 0.44, N = 3 SE +/- 1.00, N = 3 336.19 337.08 337.78 335.10
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a b c d 20 40 60 80 100 SE +/- 0.08, N = 3 SE +/- 0.12, N = 3 SE +/- 0.28, N = 3 95.02 94.88 94.68 95.45
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a b c d 13 26 39 52 65 SE +/- 0.15, N = 3 SE +/- 0.19, N = 3 SE +/- 0.19, N = 3 56.79 59.01 57.18 56.86
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a b c d 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 17.60 16.94 17.48 17.58
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a b c d 10 20 30 40 50 SE +/- 0.30, N = 3 SE +/- 0.28, N = 3 SE +/- 0.06, N = 3 43.19 43.84 43.05 43.06
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a b c d 160 320 480 640 800 SE +/- 6.44, N = 3 SE +/- 4.42, N = 3 SE +/- 2.01, N = 3 736.77 726.25 737.98 737.85
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b c d 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 17.07 17.99 16.99 17.09
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b c d 13 26 39 52 65 SE +/- 0.38, N = 3 SE +/- 0.30, N = 3 SE +/- 0.11, N = 3 58.55 55.56 58.85 58.50
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 110 220 330 440 550 SE +/- 0.37, N = 3 SE +/- 0.45, N = 3 SE +/- 0.62, N = 3 503.41 508.59 506.13 504.84
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b c d 14 28 42 56 70 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 63.48 62.81 63.13 63.30
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 14 28 42 56 70 SE +/- 0.66, N = 3 SE +/- 0.51, N = 3 SE +/- 0.57, N = 3 64.71 64.40 64.12 64.08
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b c d 4 8 12 16 20 SE +/- 0.16, N = 3 SE +/- 0.12, N = 3 SE +/- 0.14, N = 3 15.45 15.52 15.59 15.60
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a b c d 9 18 27 36 45 SE +/- 0.10, N = 3 SE +/- 0.25, N = 3 SE +/- 0.10, N = 3 37.74 37.95 37.94 37.45
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a b c d 200 400 600 800 1000 SE +/- 2.31, N = 3 SE +/- 6.31, N = 3 SE +/- 2.60, N = 3 841.43 837.46 837.14 845.44
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b c d 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.14, N = 3 SE +/- 0.03, N = 3 14.08 14.31 14.10 14.00
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b c d 16 32 48 64 80 SE +/- 0.43, N = 3 SE +/- 0.72, N = 3 SE +/- 0.16, N = 3 71.04 69.88 70.90 71.42
a Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aJava Notes: OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10)Python Notes: Python 3.11.6Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 11 December 2023 22:18 by user pts.
b Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aJava Notes: OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10)Python Notes: Python 3.11.6Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 December 2023 01:08 by user pts.
c Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aJava Notes: OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10)Python Notes: Python 3.11.6Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 December 2023 02:31 by user pts.
d Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS), Chipset: AMD Starship/Matisse, Memory: 128GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 23.10, Kernel: 6.5.0-13-generic (x86_64), Desktop: GNOME Shell 45.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.2.1-1ubuntu3 (LLVM 15.0.7 DRM 3.54), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aJava Notes: OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10)Python Notes: Python 3.11.6Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 December 2023 10:46 by user pts.