Benchmarks by Michael Larabel for a future article looking at AMD Inception impact.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2308112-NE-EPYC7763124
AMD EPYC 7763 1P spec_rstack_overflow
Benchmarks by Michael Larabel for a future article looking at AMD Inception impact.
,,"off","safe RET no microcode","safe RET","IBPB"
Processor,,AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads),AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads),AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads),AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads)
Motherboard,,AMD DAYTONA_X (RYM1009B BIOS),AMD DAYTONA_X (RYM1009B BIOS),AMD DAYTONA_X (RYM1009B BIOS),AMD DAYTONA_X (RYM1009B BIOS)
Chipset,,AMD Starship/Matisse,AMD Starship/Matisse,AMD Starship/Matisse,AMD Starship/Matisse
Memory,,256GB,256GB,256GB,256GB
Disk,,800GB INTEL SSDPF21Q800GB,800GB INTEL SSDPF21Q800GB,800GB INTEL SSDPF21Q800GB,800GB INTEL SSDPF21Q800GB
Graphics,,ASPEED,ASPEED,ASPEED,ASPEED
Monitor,,VE228,VE228,VE228,VE228
Network,,2 x Mellanox MT27710,2 x Mellanox MT27710,2 x Mellanox MT27710,2 x Mellanox MT27710
OS,,Ubuntu 22.04,Ubuntu 22.04,Ubuntu 22.04,Ubuntu 22.04
Kernel,,6.5.0-rc5-phx-tues (x86_64),6.5.0-rc5-phx-tues (x86_64),6.5.0-rc5-phx-tues (x86_64),6.5.0-rc5-phx-tues (x86_64)
Desktop,,GNOME Shell 42.5,GNOME Shell 42.5,GNOME Shell 42.5,GNOME Shell 42.5
Display Server,,X Server 1.21.1.3,X Server 1.21.1.3,X Server 1.21.1.3,X Server 1.21.1.3
Vulkan,,1.3.224,1.3.224,1.3.224,1.3.224
Compiler,,GCC 11.3.0 + LLVM 14.0.0,GCC 11.3.0 + LLVM 14.0.0,GCC 11.3.0 + LLVM 14.0.0,GCC 11.3.0 + LLVM 14.0.0
File-System,,ext4,ext4,ext4,ext4
Screen Resolution,,1920x1080,1920x1080,1920x1080,1920x1080
,,"off","safe RET no microcode","safe RET","IBPB"
"OpenVKL - Benchmark: vklBenchmark ISPC (Items / Sec)",HIB,453,452,453,450
"TensorFlow - Device: CPU - Batch Size: 64 - Model: ResNet-50 (images/sec)",HIB,17.78,15.56,15.65,17.45
"MariaDB - Clients: 8192 (Queries/sec)",HIB,355,301,301,276
"Timed Linux Kernel Compilation - Build: allmodconfig (sec)",LIB,289.063,344.242,338.157,352.178
"CockroachDB - Workload: KV, 50% Reads - Concurrency: 128 (ops/s)",HIB,103635.0,100851.4,99601.6,95416.0
"CockroachDB - Workload: KV, 95% Reads - Concurrency: 128 (ops/s)",HIB,135187.2,131487.0,132046.0,119163.8
"ClickHouse - 100M Rows Hits Dataset, Third Run (Queries/min, Geo Mean)",HIB,362.64,329.19,337.45,349.50
"ClickHouse - 100M Rows Hits Dataset, Second Run (Queries/min, Geo Mean)",HIB,361.81,337.01,337.12,347.92
"ClickHouse - 100M Rows Hits Dataset, First Run / Cold Cache (Queries/min, Geo Mean)",HIB,349.43,323.42,318.12,336.69
"MariaDB - Clients: 4096 (Queries/sec)",HIB,590,412,418,274
"OpenFOAM - Input: drivaerFastback, Medium Mesh Size - Execution Time (sec)",LIB,633.51902,644.36223,643.71316,645.40958
"OpenFOAM - Input: drivaerFastback, Medium Mesh Size - Mesh Time (sec)",LIB,140.61562,145.06069,144.02174,148.25442
"Timed LLVM Compilation - Build System: Ninja (sec)",LIB,176.374,182.169,181.528,204.080
"OSPRay - Benchmark: particle_volume/pathtracer/real_time (Items/sec)",HIB,157.829,155.165,156.419,153.439
"Timed Node.js Compilation - Time To Compile (sec)",LIB,164.268,172.749,173.064,195.493
"OpenRadioss - Model: INIVOL and Fluid Structure Interaction Drop Container (sec)",LIB,162.13,163.02,163.97,171.75
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500 (Latency)",LIB,117.97,123.61,120.06,120.53
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500 (point/sec)",HIB,39463981.42,37720117.40,38833415.97,38572529.56
"Numpy Benchmark - (Score)",HIB,457.23,418.95,422.58,389.92
"OpenRadioss - Model: Bird Strike on Windshield (sec)",LIB,144.83,152.91,152.27,160.70
"OSPRay - Benchmark: particle_volume/scivis/real_time (Items/sec)",HIB,17.7511,17.7305,17.7305,17.6743
"Neural Magic DeepSparse - Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,678.9330,679.5766,682.2071,681.3961
"Neural Magic DeepSparse - Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,46.7020,46.7239,46.5677,46.6121
"PostgreSQL - Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency (ms)",LIB,0.256,0.296,0.289,0.461
"PostgreSQL - Scaling Factor: 100 - Clients: 800 - Mode: Read Only (TPS)",HIB,3128719,2707280,2768445,1733827
"PostgreSQL - Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency (ms)",LIB,12.988,14.499,14.589,15.854
"PostgreSQL - Scaling Factor: 100 - Clients: 800 - Mode: Read Write (TPS)",HIB,61604,55175,54837,50463
"Timed MrBayes Analysis - Primate Phylogeny Analysis (sec)",LIB,136.686,137.518,138.851,144.829
"Apache Cassandra - Test: Writes (Op/s)",HIB,238741,233069,236241,220814
"Timed Godot Game Engine Compilation - Time To Compile (sec)",LIB,121.948,125.663,125.060,135.962
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time (sec)",LIB,1.30,1.39,1.41,1.64
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Inner Join Test Time (sec)",LIB,1.88,2.22,2.14,2.41
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Repartition Test Time (sec)",LIB,2.09,2.38,2.26,2.31
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Group By Test Time (sec)",LIB,4.91,5.17,5.15,5.74
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark (sec)",LIB,31.84,32.02,31.43,31.49
"Apache Spark - Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time (sec)",LIB,3.39,3.42,3.47,3.71
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 (Latency)",LIB,78.92,79.19,82.24,80.78
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 (point/sec)",HIB,58682618.18,58073516.79,57099408.15,57529201.42
"Redis 7.0.12 + memtier_benchmark - Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10 (Ops/sec)",HIB,2195705.51,2154339.26,2157815.69,2148876.03
"OpenRadioss - Model: Bumper Beam (sec)",LIB,87.72,93.68,93.90,113.68
"OSPRay - Benchmark: particle_volume/ao/real_time (Items/sec)",HIB,18.0226,18.0288,17.9817,17.9663
"OpenRadioss - Model: Rubber O-Ring Seal Installation (sec)",LIB,77.46,85.04,84.48,99.04
"nginx - Connections: 1000 (Reqs/sec)",HIB,166499.89,140555.98,143271.26,135431.46
"nginx - Connections: 500 (Reqs/sec)",HIB,169583.15,144020.03,142619.84,137051.69
"Neural Magic DeepSparse - Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,55.3922,55.3747,55.4028,55.5026
"Neural Magic DeepSparse - Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,576.9722,577.0589,576.8166,575.7335
"Blender - Blend File: Pabellon Barcelona - Compute: CPU-Only (sec)",LIB,84.50,84.69,84.49,85.63
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200 (Latency)",LIB,37.70,35.10,37.53,36.47
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200 (point/sec)",HIB,43665846.28,46538766.01,44027904.89,44816394.74
"Timed Linux Kernel Compilation - Build: defconfig (sec)",LIB,31.192,37.623,37.243,40.085
"OpenVINO - Model: Person Detection FP16 - Device: CPU (ms)",LIB,4092.91,4124.74,4114.58,4204.34
"OpenVINO - Model: Person Detection FP16 - Device: CPU (FPS)",HIB,7.68,7.58,7.60,7.46
"Redis 7.0.12 + memtier_benchmark - Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:5 (Ops/sec)",HIB,2197287.30,2167181.09,2145052.14,2126493.29
"Redis 7.0.12 + memtier_benchmark - Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:5 (Ops/sec)",HIB,2204628.92,2218601.79,2145436.26,2092844.22
"Redis 7.0.12 + memtier_benchmark - Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10 (Ops/sec)",HIB,2177211.80,2173694.77,2172804.71,2137964.98
"OpenVINO - Model: Face Detection FP16-INT8 - Device: CPU (ms)",LIB,1141.43,1142.29,1142.53,1144.98
"OpenVINO - Model: Face Detection FP16-INT8 - Device: CPU (FPS)",HIB,27.83,27.79,27.82,27.71
"RocksDB - Test: Read Random Write Random (Op/s)",HIB,2951684,2872765,2839085,2130006
"OSPRay - Benchmark: gravity_spheres_volume/dim_512/scivis/real_time (Items/sec)",HIB,8.33174,8.32749,8.33813,8.26549
"OSPRay - Benchmark: gravity_spheres_volume/dim_512/ao/real_time (Items/sec)",HIB,8.96051,8.96941,8.94049,8.89239
"Apache Spark - Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe (sec)",LIB,,,,2.60
"OpenVINO - Model: Weld Porosity Detection FP16 - Device: CPU (ms)",LIB,28.40,28.38,28.39,28.62
"OpenVINO - Model: Weld Porosity Detection FP16 - Device: CPU (FPS)",HIB,1126.03,1126.64,1126.14,1116.89
"Neural Magic DeepSparse - Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,65.5911,65.8910,65.5662,65.7767
"Neural Magic DeepSparse - Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,487.2450,485.0069,487.3677,485.6842
"RocksDB - Test: Update Random (Op/s)",HIB,462287,428112,426947,322231
"Neural Magic DeepSparse - Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,596.5915,596.5579,596.7822,596.7902
"Neural Magic DeepSparse - Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,53.5968,53.5972,53.5729,53.5782
"OSPRay - Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time (Items/sec)",HIB,13.1355,13.2621,13.2538,13.1709
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200 (Latency)",LIB,36.58,38.54,35.82,36.56
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200 (point/sec)",HIB,49501499.13,47445770.18,50578426.54,49316970.28
"Neural Magic DeepSparse - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,839.7201,840.7721,840.4236,840.8881
"Neural Magic DeepSparse - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,37.6037,37.6987,37.6319,37.6818
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200 (Latency)",LIB,13.83,14.05,14.73,14.78
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200 (point/sec)",HIB,960525.66,947741.34,918691.45,921701.44
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500 (Latency)",LIB,31.49,31.93,27.73,31.11
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500 (point/sec)",HIB,1415756.33,1408658.83,1583717.62,1441637.63
"OpenRadioss - Model: Cell Phone Drop Test (sec)",LIB,33.10,36.37,36.40,40.11
"GROMACS - Implementation: MPI CPU - Input: water_GMX50_bare (Ns/Day)",HIB,5.680,5.730,5.706,5.707
"Neural Magic DeepSparse - Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,68.1483,68.2595,68.2325,68.3456
"Neural Magic DeepSparse - Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,468.8306,468.3646,468.2170,467.3150
"7-Zip Compression - Test: Decompression Rating (MIPS)",HIB,385585,383039,383515,385487
"7-Zip Compression - Test: Compression Rating (MIPS)",HIB,384374,334812,335595,371799
"Algebraic Multi-Grid Benchmark - (Figure Of Merit)",HIB,1011799000,999645400,999102100,1005138667
"Neural Magic DeepSparse - Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream (ms/batch)",LIB,8.3078,8.3612,8.3219,8.3411
"Neural Magic DeepSparse - Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream (items/sec)",HIB,3840.6307,3816.7676,3834.2439,3824.8431
"Remhos - Test: Sample Remap Example (sec)",LIB,17.375,17.788,17.958,18.658
"SPECFEM3D - Model: Layered Halfspace (sec)",LIB,31.845630424,31.829261870,31.659940885,32.009762142
"SPECFEM3D - Model: Water-layered Halfspace (sec)",LIB,29.772535386,30.427531709,29.590868260,30.053834808
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500 (Latency)",LIB,32.36,30.29,30.14,31.74
"Apache IoTDB - Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500 (point/sec)",HIB,1271946.57,1342031.11,1345598.59,1287324.35
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200 (Latency)",LIB,14.05,13.61,13.36,11.93
"Apache IoTDB - Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200 (point/sec)",HIB,1176385.35,1202637.36,1211172.13,1344749.20
"NAMD - ATPase Simulation - 327,506 Atoms (days/ns)",LIB,0.38130,0.38115,0.38098,0.38534
"Blender - Blend File: BMW27 - Compute: CPU-Only (sec)",LIB,27.34,27.58,27.46,27.73
"SPECFEM3D - Model: Homogeneous Halfspace (sec)",LIB,17.417120933,17.643680397,17.690298650,17.695914319
"SPECFEM3D - Model: Tomographic Model (sec)",LIB,14.134265606,14.404419238,14.188058962,14.225825227
"DaCapo Benchmark - Java Test: Jython (msec)",LIB,4193,4191,4241,4446
"ACES DGEMM - Sustained Floating-Point Rate (GFLOP/s)",HIB,24.200551,24.695818,23.702889,24.251474
"DaCapo Benchmark - Java Test: Tradebeans (msec)",LIB,3993,4096,4143,5305
"SPECFEM3D - Model: Mount St. Helens (sec)",LIB,11.801238732,11.982163460,12.010380781,11.951448577
"Embree - Binary: Pathtracer ISPC - Model: Crown (FPS)",HIB,57.4229,57.2956,57.3138,57.2967
"Embree - Binary: Pathtracer ISPC - Model: Asian Dragon (FPS)",HIB,64.5964,64.3916,64.6742,63.4668
"SQLite - Threads / Copies: 16 (sec)",LIB,6.273,8.834,8.800,7.934
"SQLite - Threads / Copies: 8 (sec)",LIB,3.755,4.850,5.006,4.793