Microsoft Azure EPYC 7003 HBv3 Benchmarks

Azure HBv3 benchmarks against other Microsoft Azure HPC instance types. Benchmarks by Michael Larabel for a future article on phoronix.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2104126-PTS-MSAZURE208
Jump To Table - Results

View

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

Limit displaying results to tests within:

Timed Code Compilation 3 Tests
C/C++ Compiler Tests 6 Tests
CPU Massive 15 Tests
Creator Workloads 4 Tests
Encoding 3 Tests
Finance 2 Tests
Fortran Tests 4 Tests
HPC - High Performance Computing 15 Tests
Machine Learning 5 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 5 Tests
Multi-Core 14 Tests
NVIDIA GPU Compute 4 Tests
OpenMPI Tests 8 Tests
Programmer / Developer System Benchmarks 4 Tests
Python Tests 3 Tests
Scientific Computing 7 Tests
Server CPU Tests 10 Tests
Single-Threaded 2 Tests
Video Encoding 3 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Azure HBv3 - EPYC 7V13
April 09 2021
  7 Hours, 26 Minutes
Azure HBv2 - EPYC 7V12
April 10 2021
  7 Hours, 35 Minutes
Azure HBv1 - EPYC 7551
April 11 2021
  9 Hours, 54 Minutes
Azure HC - Xeon 8168
April 12 2021
  4 Hours, 27 Minutes
Invert Hiding All Results Option
  7 Hours, 20 Minutes

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


Microsoft Azure EPYC 7003 HBv3 Benchmarks Azure HBv3 benchmarks against other Microsoft Azure HPC instance types. Benchmarks by Michael Larabel for a future article on phoronix. ,,"Azure HBv3 - EPYC 7V13","Azure HBv2 - EPYC 7V12","Azure HBv1 - EPYC 7551","Azure HC - Xeon 8168" Processor,,2 x AMD EPYC 7V13 64-Core (120 Cores),2 x AMD EPYC 7V12 64-Core (120 Cores),2 x AMD EPYC 7551 32-Core (60 Cores),2 x Intel Xeon Platinum 8168 (44 Cores) Motherboard,,Microsoft Virtual Machine (Hyper-V UEFI v4.1 BIOS),Microsoft Virtual Machine (Hyper-V UEFI v4.0 BIOS),Microsoft Virtual Machine (Hyper-V UEFI v4.0 BIOS),Microsoft Virtual Machine (Hyper-V UEFI v4.0 BIOS) Memory,,442GB,450GB,226GB,348GB Disk,,2 x 960GB Microsoft NVMe Direct Disk + 32GB Virtual Disk + 515GB Virtual Disk,960GB Microsoft NVMe Direct Disk + 32GB Virtual Disk + 515GB Virtual Disk,32GB Virtual Disk + 752GB Virtual Disk,32GB Virtual Disk + 752GB Virtual Disk Graphics,,hyperv_fb,hyperv_fb,hyperv_fb,hyperv_fb OS,,CentOS Linux 8,CentOS Linux 8,CentOS Linux 8,CentOS Linux 8 Kernel,,4.18.0-147.8.1.el8_1.x86_64 (x86_64),4.18.0-147.8.1.el8_1.x86_64 (x86_64),4.18.0-147.8.1.el8_1.x86_64 (x86_64),4.18.0-147.8.1.el8_1.x86_64 (x86_64) Compiler,,GCC 8.3.1 20190507,GCC 8.3.1 20190507,GCC 8.3.1 20190507,GCC 8.3.1 20190507 File-System,,nfs,nfs,nfs,nfs Screen Resolution,,1152x864,1152x864,1152x864,1152x864 System Layer,,microsoft,microsoft,microsoft,microsoft ,,"Azure HBv3 - EPYC 7V13","Azure HBv2 - EPYC 7V12","Azure HBv1 - EPYC 7551","Azure HC - Xeon 8168" "Botan - Test: KASUMI (MiB/s)",HIB,87.480,73.175,57.950,76.149 "Botan - Test: KASUMI - Decrypt (MiB/s)",HIB,84.059,71.101,55.980,74.287 "Botan - Test: AES-256 (MiB/s)",HIB,5412.131,3934.401,3144.313,3155.391 "Botan - Test: AES-256 - Decrypt (MiB/s)",HIB,5407.477,3938.444,3151.228,3159.852 "Botan - Test: Twofish (MiB/s)",HIB,346.319,280.902,217.312,293.551 "Botan - Test: Twofish - Decrypt (MiB/s)",HIB,345.776,283.257,218.034,294.381 "Botan - Test: Blowfish (MiB/s)",HIB,424.738,348.425,273.039,367.942 "Botan - Test: Blowfish - Decrypt (MiB/s)",HIB,425.158,349.147,272.733,365.547 "Botan - Test: CAST-256 (MiB/s)",HIB,133.691,113.600,88.545,116.630 "Botan - Test: CAST-256 - Decrypt (MiB/s)",HIB,133.665,113.724,88.555,116.573 "Botan - Test: ChaCha20Poly1305 (MiB/s)",HIB,667.765,615.795,303.243,633.217 "Botan - Test: ChaCha20Poly1305 - Decrypt (MiB/s)",HIB,658.370,610.861,300.178,626.919 "QuantLib - (MFLOPS)",HIB,2299.3,1725.2,1251.7,1812.9 "Timed MAFFT Alignment - Multiple Sequence Alignment - LSU RNA (sec)",LIB,14.326,16.064,22.067,15.627 "PlaidML - FP16: No - Mode: Inference - Network: VGG16 - Device: CPU (FPS)",HIB,38.39,25.57,19.88,41.54 "PlaidML - FP16: No - Mode: Inference - Network: VGG19 - Device: CPU (FPS)",HIB,34.53,23.12,15.97,34.66 "PlaidML - FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU (FPS)",HIB,6.27,5.60,4.80,4.91 "CloverLeaf - Lagrangian-Eulerian Hydrodynamics (sec)",LIB,16.66,23.78,25.00,20.19 "Mobile Neural Network - Model: SqueezeNetV1.0 (ms)",LIB,8.224,14.136,17.194,8.226 "Mobile Neural Network - Model: resnet-v2-50 (ms)",LIB,29.086,52.533,62.446,29.006 "Mobile Neural Network - Model: inception-v3 (ms)",LIB,34.710,55.356,76.799,35.673 "TensorFlow Lite - Model: SqueezeNet (us)",LIB,66320.4,74885.4,95206.2,73517.3 "TNN - Target: CPU - Model: SqueezeNet v1.1 (ms)",LIB,272.614,323.615,420.047,398.404 "GROMACS - Water Benchmark (Ns/Day)",HIB,9.041,8.458,3.288,4.866 "High Performance Conjugate Gradient - (GFLOP/s)",HIB,39.0620,37.2686,30.4314,26.0381 "NAS Parallel Benchmarks - Test / Class: LU.C (Mop/s)",HIB,56682.82,53829.34,25304.19, "Rodinia - Test: OpenMP LavaMD (sec)",LIB,37.803,39.198,90.175,94.119 "Rodinia - Test: OpenMP HotSpot3D (sec)",LIB,82.489,107.976,145.569,115.832 "NAMD - ATPase Simulation - 327,506 Atoms (days/ns)",LIB,0.27566,0.30056,0.70947,0.52569 "oneDNN - Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU (ms)",LIB,0.881307,1.12484,1.87834,0.931073 "oneDNN - Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU (ms)",LIB,3.21204,9.68321,10.22263,3.28160 "oneDNN - Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,0.383855,0.441012,1.76355,0.474311 "oneDNN - Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU (ms)",LIB,9.50445,12.1051,7.91569,2.67914 "oneDNN - Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU (ms)",LIB,1.58182,1.59557,3.94077,1.21181 "oneDNN - Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,0.406686,1.031942,2.40561,0.551467 "oneDNN - Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,0.455682,0.732130,2.73621,0.668641 "oneDNN - Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU (ms)",LIB,875.166,1314.00,5524.50,744.355 "oneDNN - Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU (ms)",LIB,530.898,778.305,3344.14,462.855 "oneDNN - Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,962.336,1287.03,5539.48,746.464 "oneDNN - Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,561.773,791.460,3373.88,459.576 "oneDNN - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU (ms)",LIB,0.255501,0.350347,1.51271,0.352180 "oneDNN - Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU (ms)",LIB,843.103,1294.53,5355.24,753.356 "oneDNN - Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU (ms)",LIB,540.194,796.816,3423.15,458.201 "oneDNN - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU (ms)",LIB,0.445223,0.524737,1.15465,0.398005 "Pennant - Test: sedovbig (Hydro Cycle Time - sec)",LIB,5.833419,6.026449,15.52899,23.87126 "Pennant - Test: leblancbig (Hydro Cycle Time - sec)",LIB,3.337201,3.486503,10.33764,10.52920 "LULESH - (z/s)",HIB,41636.603,34803.539,19011.458,20846.590 "Xcompact3d Incompact3d - Input: X3D-benchmarking input.i3d (sec)",LIB,287.599508,318.905009,455.812378,507.295970 "Timed LLVM Compilation - Time To Compile (sec)",LIB,163.760,174.743,283.291,251.744 "Zstd Compression - Compression Level: 8 - Compression Speed (MB/s)",HIB,3184.2,2653.1,920.1,1188.2 "Zstd Compression - Compression Level: 19 - Compression Speed (MB/s)",HIB,78.2,69.5,48.5,76.3 "Zstd Compression - Compression Level: 19 - Decompression Speed (MB/s)",HIB,3717.5,2631.0,2005.4,1903.9 "Zstd Compression - Compression Level: 8, Long Mode - Compression Speed (MB/s)",HIB,771.9,588.0,468.8,649.3 "Zstd Compression - Compression Level: 8, Long Mode - Decompression Speed (MB/s)",HIB,4256.8,3037.5,2339.6,2350.2 "Zstd Compression - Compression Level: 19, Long Mode - Decompression Speed (MB/s)",HIB,3727.5,2674.2,2019.3,1991.3 "Timed Linux Kernel Compilation - Time To Compile (sec)",LIB,42.008,44.922,58.738,55.324 "SVT-VP9 - Tuning: Visual Quality Optimized - Input: Bosphorus 1080p (FPS)",HIB,337.71,140.59,129.44,299.95 "SVT-AV1 - Encoder Mode: Enc Mode 0 - Input: 1080p (FPS)",HIB,0.161,0.124,0.076,0.111 "SVT-AV1 - Encoder Mode: Enc Mode 4 - Input: 1080p (FPS)",HIB,12.275,9.512,4.669,7.621 "SVT-AV1 - Encoder Mode: Enc Mode 8 - Input: 1080p (FPS)",HIB,94.028,78.104,40.146,57.377 "SVT-HEVC - Tuning: 1 - Input: Bosphorus 1080p (FPS)",HIB,45.39,36.12,21.12,23.87 "SVT-HEVC - Tuning: 7 - Input: Bosphorus 1080p (FPS)",HIB,378.01,166.90,119.20,256.28 "SVT-HEVC - Tuning: 10 - Input: Bosphorus 1080p (FPS)",HIB,548.33,379.91,141.93,480.79 "Timed Node.js Compilation - Time To Compile (sec)",LIB,111.327,118.969,194.550,189.096 "FinanceBench - Benchmark: Repo OpenMP (ms)",LIB,59196.546875,75154.257813,108606.502604,71609.065104 "FinanceBench - Benchmark: Bonds OpenMP (ms)",LIB,104884.294271,133719.328125,195702.796875,130250.449219 "GNU GMP GMPbench - Total Time (GMPbench Score)",HIB,4893.4,4155.5,3021.3,4196.5