7f32 feb AMD EPYC 7F32 8-Core testing with a ASRockRack EPYCD8 (P2.40 BIOS) and ASPEED on Debian 11 via the Phoronix Test Suite. a: Processor: AMD EPYC 7F32 8-Core @ 3.70GHz (8 Cores / 16 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 28GB, Disk: Samsung SSD 970 EVO Plus 250GB, Graphics: ASPEED, Network: 2 x Intel I350 OS: Debian 11, Kernel: 5.10.0-10-amd64 (x86_64), Desktop: GNOME Shell 3.38.6, Display Server: X Server, Compiler: GCC 10.2.1 20210110, File-System: ext4, Screen Resolution: 1024x768 b: Processor: AMD EPYC 7F32 8-Core @ 3.70GHz (8 Cores / 16 Threads), Motherboard: ASRockRack EPYCD8 (P2.40 BIOS), Chipset: AMD Starship/Matisse, Memory: 28GB, Disk: Samsung SSD 970 EVO Plus 250GB, Graphics: ASPEED, Network: 2 x Intel I350 OS: Debian 11, Kernel: 5.10.0-10-amd64 (x86_64), Desktop: GNOME Shell 3.38.6, Display Server: X Server, Compiler: GCC 10.2.1 20210110, File-System: ext4, Screen Resolution: 1024x768 dav1d 1.1 Video Input: Chimera 1080p FPS > Higher Is Better a . 390.36 |=================================================================== b . 390.37 |=================================================================== dav1d 1.1 Video Input: Summer Nature 4K FPS > Higher Is Better a . 170.85 |================================================================ b . 178.99 |=================================================================== dav1d 1.1 Video Input: Summer Nature 1080p FPS > Higher Is Better a . 612.98 |=================================================================== b . 616.71 |=================================================================== dav1d 1.1 Video Input: Chimera 1080p 10-bit FPS > Higher Is Better a . 366.43 |================================================================== b . 369.21 |=================================================================== AOM AV1 3.6 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 0.19 |===================================================================== b . 0.19 |===================================================================== AOM AV1 3.6 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 5.64 |===================================================================== b . 5.65 |===================================================================== AOM AV1 3.6 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 45.97 |=================================================================== b . 46.60 |==================================================================== AOM AV1 3.6 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 9.56 |===================================================================== b . 9.53 |===================================================================== AOM AV1 3.6 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 37.74 |=================================================================== b . 38.13 |==================================================================== AOM AV1 3.6 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 45.36 |================================================================ b . 48.38 |==================================================================== AOM AV1 3.6 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 44.60 |================================================================== b . 46.21 |==================================================================== AOM AV1 3.6 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 0.54 |===================================================================== b . 0.54 |===================================================================== AOM AV1 3.6 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 12.30 |==================================================================== b . 12.35 |==================================================================== AOM AV1 3.6 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 107.42 |=================================================================== b . 100.54 |=============================================================== AOM AV1 3.6 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 28.80 |==================================================================== b . 28.22 |=================================================================== AOM AV1 3.6 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 101.40 |=================================================================== b . 95.22 |=============================================================== AOM AV1 3.6 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 114.06 |================================================================== b . 115.24 |=================================================================== AOM AV1 3.6 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 125.49 |================================================================== b . 126.50 |=================================================================== Embree 4.0 Binary: Pathtracer - Model: Crown Frames Per Second > Higher Is Better a . 9.5705 |================================================================= b . 9.8828 |=================================================================== Embree 4.0 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better a . 9.3491 |=================================================================== b . 9.3296 |=================================================================== Embree 4.0 Binary: Pathtracer - Model: Asian Dragon Frames Per Second > Higher Is Better a . 10.54 |================================================================== b . 10.81 |==================================================================== Embree 4.0 Binary: Pathtracer - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 9.7218 |=================================================================== b . 9.7600 |=================================================================== Embree 4.0 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better a . 10.88 |==================================================================== b . 10.79 |=================================================================== Embree 4.0 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 9.3827 |=================================================================== b . 9.3503 |=================================================================== VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 5.42 |===================================================================== b . 5.44 |===================================================================== VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 9.80 |===================================================================== b . 9.61 |==================================================================== VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 11.26 |==================================================================== b . 11.01 |================================================================== VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 23.93 |==================================================================== b . 23.25 |================================================================== VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Fast Frames Per Second > Higher Is Better a . 2.72 |===================================================================== b . 2.71 |===================================================================== VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Faster Frames Per Second > Higher Is Better a . 6.274 |==================================================================== b . 6.085 |================================================================== VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Fast Frames Per Second > Higher Is Better a . 7.229 |==================================================================== b . 7.257 |==================================================================== VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Faster Frames Per Second > Higher Is Better a . 17.60 |==================================================================== b . 17.60 |==================================================================== ClickHouse 22.12.3.5 Queries Per Minute, Geo Mean > Higher Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time Seconds < Lower Is Better a . 4.03 |===================================================================== b . 4.04 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Seconds < Lower Is Better a . 188.34 |=================================================================== b . 187.45 |=================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe Seconds < Lower Is Better a . 11.49 |==================================================================== b . 11.52 |==================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Group By Test Time Seconds < Lower Is Better a . 4.64 |===================================================================== b . 4.59 |==================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Repartition Test Time Seconds < Lower Is Better a . 2.00 |==================================================================== b . 2.02 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Inner Join Test Time Seconds < Lower Is Better a . 2.13 |===================================================================== b . 2.05 |================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time Seconds < Lower Is Better a . 1.93 |===================================================================== b . 1.80 |================================================================ Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - SHA-512 Benchmark Time Seconds < Lower Is Better a . 4.15 |===================================================================== b . 4.13 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Calculate Pi Benchmark Seconds < Lower Is Better a . 189.28 |=================================================================== b . 189.00 |=================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Calculate Pi Benchmark Using Dataframe Seconds < Lower Is Better a . 11.39 |================================================================== b . 11.69 |==================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Group By Test Time Seconds < Lower Is Better a . 5.05 |===================================================================== b . 5.02 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Repartition Test Time Seconds < Lower Is Better a . 1.96 |============================================================= b . 2.23 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Inner Join Test Time Seconds < Lower Is Better a . 2.30 |================================================================ b . 2.49 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 500 - Broadcast Inner Join Test Time Seconds < Lower Is Better a . 2.07 |===================================================================== b . 1.91 |================================================================ Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - SHA-512 Benchmark Time Seconds < Lower Is Better a . 4.32 |===================================================================== b . 4.29 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Calculate Pi Benchmark Seconds < Lower Is Better a . 186.61 |=================================================================== b . 185.93 |=================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Calculate Pi Benchmark Using Dataframe Seconds < Lower Is Better a . 11.43 |=================================================================== b . 11.54 |==================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Group By Test Time Seconds < Lower Is Better a . 5.59 |===================================================================== b . 5.31 |================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Repartition Test Time Seconds < Lower Is Better a . 2.25 |===================================================================== b . 2.22 |==================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Inner Join Test Time Seconds < Lower Is Better a . 2.77 |===================================================================== b . 2.43 |============================================================= Apache Spark 3.3 Row Count: 1000000 - Partitions: 1000 - Broadcast Inner Join Test Time Seconds < Lower Is Better a . 2.17651277 |========================================================== b . 2.38000000 |=============================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - SHA-512 Benchmark Time Seconds < Lower Is Better a . 4.74 |==================================================================== b . 4.78 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Calculate Pi Benchmark Seconds < Lower Is Better a . 186.58 |=================================================================== b . 186.16 |=================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Calculate Pi Benchmark Using Dataframe Seconds < Lower Is Better a . 11.51 |==================================================================== b . 11.40 |=================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Group By Test Time Seconds < Lower Is Better a . 5.79 |===================================================================== b . 5.77 |===================================================================== Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Repartition Test Time Seconds < Lower Is Better a . 2.753324538 |============================================================== b . 2.690000000 |============================================================= Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Inner Join Test Time Seconds < Lower Is Better a . 3.18 |===================================================================== b . 3.01 |================================================================= Apache Spark 3.3 Row Count: 1000000 - Partitions: 2000 - Broadcast Inner Join Test Time Seconds < Lower Is Better a . 2.39 |============================================================= b . 2.72 |===================================================================== GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better a . 1.244 |================================================================== b . 1.283 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1 - Mode: Read Only TPS > Higher Is Better a . 61593 |================================================================= b . 64239 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.016 |==================================================================== b . 0.016 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1 - Mode: Read Write TPS > Higher Is Better a . 708 |====================================================================== b . 703 |====================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average Latency ms < Lower Is Better a . 1.413 |==================================================================== b . 1.423 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Only TPS > Higher Is Better a . 620983 |=================================================================== b . 618290 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.081 |==================================================================== b . 0.081 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 100 - Mode: Read Only TPS > Higher Is Better a . 627441 |================================================================= b . 645705 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.159 |==================================================================== b . 0.155 |================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 250 - Mode: Read Only TPS > Higher Is Better a . 606075 |================================================================== b . 617509 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 250 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.412 |==================================================================== b . 0.405 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Write TPS > Higher Is Better a . 678 |====================================================================== b . 675 |====================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency ms < Lower Is Better a . 73.72 |==================================================================== b . 74.05 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 500 - Mode: Read Only TPS > Higher Is Better a . 522138 |============================================================= b . 571774 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 500 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.958 |==================================================================== b . 0.874 |============================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Only TPS > Higher Is Better a . 499392 |=================================================================== b . 479473 |================================================================ PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.602 |================================================================= b . 1.668 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1 - Mode: Read Only TPS > Higher Is Better a . 54842 |=================================================================== b . 55389 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.018 |==================================================================== b . 0.018 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 100 - Mode: Read Write TPS > Higher Is Better a . 658 |====================================================================== b . 654 |====================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average Latency ms < Lower Is Better a . 152.08 |=================================================================== b . 152.86 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better a . 466800 |=================================================================== b . 450620 |================================================================= PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better a . 2.142 |================================================================== b . 2.219 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 250 - Mode: Read Write TPS > Higher Is Better a . 598 |==================================================================== b . 614 |====================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 250 - Mode: Read Write - Average Latency ms < Lower Is Better a . 417.72 |=================================================================== b . 407.38 |================================================================= PostgreSQL 15 Scaling Factor: 1 - Clients: 500 - Mode: Read Write TPS > Higher Is Better a . 509 |====================================================================== b . 488 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 500 - Mode: Read Write - Average Latency ms < Lower Is Better a . 982.95 |=============================================================== b . 1025.32 |================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Write TPS > Higher Is Better a . 326 |===================================================================== b . 329 |====================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better a . 2457.45 |================================================================== b . 2428.66 |================================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 1 - Mode: Read Write TPS > Higher Is Better a . 425 |====================================================================== b . 424 |====================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average Latency ms < Lower Is Better a . 2.355 |==================================================================== b . 2.360 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Only TPS > Higher Is Better a . 545224 |=================================================================== b . 531797 |================================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.092 |=================================================================== b . 0.094 |==================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better a . 270 |====================================================================== b . 258 |=================================================================== PostgreSQL 15 Scaling Factor: 1 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better a . 3707.47 |=============================================================== b . 3876.57 |================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 100 - Mode: Read Only TPS > Higher Is Better a . 561050 |=================================================================== b . 563522 |=================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.178 |==================================================================== b . 0.177 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 250 - Mode: Read Only TPS > Higher Is Better a . 550537 |=================================================================== b . 543855 |================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.454 |=================================================================== b . 0.460 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Write TPS > Higher Is Better a . 4856 |===================================================================== b . 4516 |================================================================ PostgreSQL 15 Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average Latency ms < Lower Is Better a . 10.30 |=============================================================== b . 11.07 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Only TPS > Higher Is Better a . 491975 |=================================================================== b . 470230 |================================================================ PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.016 |================================================================= b . 1.063 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only TPS > Higher Is Better a . 446689 |============================================================= b . 486710 |=================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.791 |==================================================================== b . 1.644 |============================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 100 - Mode: Read Write TPS > Higher Is Better a . 6893 |===================================================================== b . 6055 |============================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latency ms < Lower Is Better a . 14.51 |============================================================ b . 16.52 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better a . 410670 |====================================================== b . 508680 |=================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better a . 2.435 |==================================================================== b . 1.966 |======================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 250 - Mode: Read Write TPS > Higher Is Better a . 8126 |===================================================================== b . 7069 |============================================================ PostgreSQL 15 Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency ms < Lower Is Better a . 30.77 |=========================================================== b . 35.37 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Write TPS > Higher Is Better a . 8425 |===================================================================== b . 6839 |======================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average Latency ms < Lower Is Better a . 59.35 |======================================================= b . 73.11 |==================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write TPS > Higher Is Better a . 8166 |===================================================================== b . 7173 |============================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better a . 97.97 |=========================================================== b . 111.52 |=================================================================== PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better a . 7792 |===================================================================== b . 6889 |============================================================= PostgreSQL 15 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better a . 128.35 |=========================================================== b . 145.16 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 6.3968 |=================================================================== b . 6.3828 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 618.83 |=================================================================== b . 619.46 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 5.4210 |=================================================================== b . 5.3209 |================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 184.46 |================================================================== b . 187.93 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 59.38 |==================================================================== b . 59.57 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 67.30 |==================================================================== b . 67.04 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 33.53 |================================================================ b . 35.64 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 29.81 |==================================================================== b . 28.05 |================================================================ Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 22.42 |=================================================================== b . 22.59 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 177.91 |=================================================================== b . 176.59 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 12.09 |================================================================ b . 12.91 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 82.73 |==================================================================== b . 77.42 |================================================================ Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 36.49 |==================================================================== b . 36.22 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 109.45 |================================================================== b . 110.31 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 26.15 |=================================================================== b . 26.37 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 38.22 |==================================================================== b . 37.91 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 72.28 |==================================================================== b . 71.54 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 55.27 |=================================================================== b . 55.85 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 52.31 |==================================================================== b . 50.75 |================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 19.11 |================================================================== b . 19.70 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 50.37 |==================================================================== b . 50.16 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 79.29 |==================================================================== b . 79.62 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 36.57 |==================================================================== b . 36.07 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 27.34 |=================================================================== b . 27.72 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 8.0865 |================================================================== b . 8.1627 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 492.13 |=================================================================== b . 487.20 |================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 6.8129 |=================================================================== b . 6.8210 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 146.76 |=================================================================== b . 146.59 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 25.55 |==================================================================== b . 25.27 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 156.34 |================================================================== b . 157.89 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 18.53 |==================================================================== b . 18.47 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 53.96 |==================================================================== b . 54.12 |==================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 6.4027 |=================================================================== b . 6.3771 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 618.22 |=================================================================== b . 618.84 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 5.3727 |=================================================================== b . 5.3372 |=================================================================== Neural Magic DeepSparse 1.3.2 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 186.12 |=================================================================== b . 187.36 |=================================================================== OpenEMS 0.0.35-86 Test: pyEMS Coupler MCells/s > Higher Is Better a . 20.99 |==================================================================== b . 20.97 |==================================================================== OpenEMS 0.0.35-86 Test: openEMS MSL_NotchFilter MCells/s > Higher Is Better a . 43.28 |================================================================== b . 44.47 |==================================================================== RocksDB 7.9.2 Test: Random Fill Op/s > Higher Is Better a . 530099 |=================================================================== b . 528550 |=================================================================== RocksDB 7.9.2 Test: Random Read Op/s > Higher Is Better a . 35559141 |================================================================ b . 36110424 |================================================================= RocksDB 7.9.2 Test: Update Random Op/s > Higher Is Better a . 346409 |=================================================================== b . 347469 |=================================================================== RocksDB 7.9.2 Test: Sequential Fill Op/s > Higher Is Better a . 589310 |================================================================== b . 595266 |=================================================================== RocksDB 7.9.2 Test: Random Fill Sync Op/s > Higher Is Better a . 2852 |===================================================================== b . 2857 |===================================================================== RocksDB 7.9.2 Test: Read While Writing Op/s > Higher Is Better a . 1490922 |================================================================== b . 1401161 |============================================================== RocksDB 7.9.2 Test: Read Random Write Random Op/s > Higher Is Better a . 1146165 |================================================================== b . 1152478 |==================================================================