Amazon AWS Graviton3E vs. Graviton 2/3 benchmarks

Benchmarks by Michael Larabel for a future article on Phoronix.com.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2308110-NE-2307106NE96
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:

BLAS (Basic Linear Algebra Sub-Routine) Tests 4 Tests
C++ Boost Tests 3 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 6 Tests
CPU Massive 10 Tests
Creator Workloads 2 Tests
Fortran Tests 6 Tests
HPC - High Performance Computing 17 Tests
Common Kernel Benchmarks 2 Tests
LAPACK (Linear Algebra Pack) Tests 2 Tests
Linear Algebra 2 Tests
Molecular Dynamics 6 Tests
MPI Benchmarks 8 Tests
Multi-Core 12 Tests
NVIDIA GPU Compute 3 Tests
OpenMPI Tests 18 Tests
Programmer / Developer System Benchmarks 5 Tests
Python Tests 5 Tests
Scientific Computing 12 Tests
Software Defined Radio 2 Tests
Server 2 Tests
Server CPU Tests 6 Tests
Common Workstation Benchmarks 2 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
m7g.16xlarge Graviton3
June 22 2023
  6 Hours, 40 Minutes
c6g.16xlarge Graviton2
June 23 2023
  8 Hours, 18 Minutes
c7g.16xlarge Graviton3
June 23 2023
  6 Hours, 40 Minutes
c7gn.16xlarge Graviton3E
July 10 2023
  6 Hours, 39 Minutes
c6a.16xlarge AMD Zen 3
August 11 2023
  8 Hours, 48 Minutes
Invert Hiding All Results Option
  7 Hours, 25 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):

Amazon AWS Graviton3E vs. Graviton 2/3 benchmarks Benchmarks by Michael Larabel for a future article on Phoronix.com. ,,"m7g.16xlarge Graviton3","c6g.16xlarge Graviton2","c7g.16xlarge Graviton3","c7gn.16xlarge Graviton3E","c6a.16xlarge AMD Zen 3" Processor,,ARMv8 Neoverse-V1 (64 Cores),ARMv8 Neoverse-N1 (64 Cores),ARMv8 Neoverse-V1 (64 Cores),ARMv8 Neoverse-V1 (64 Cores),AMD EPYC 7R13 (32 Cores / 64 Threads) Motherboard,,Amazon EC2 m7g.16xlarge (1.0 BIOS),Amazon EC2 c6g.16xlarge (1.0 BIOS),Amazon EC2 c7g.16xlarge (1.0 BIOS),Amazon EC2 c7gn.16xlarge (1.0 BIOS),Amazon EC2 c6a.16xlarge (1.0 BIOS) Chipset,,Amazon Device 0200,Amazon Device 0200,Amazon Device 0200,Amazon Device 0200,Intel 440FX 82441FX PMC Memory,,256GB,128GB,128GB,128GB,128GB Disk,,215GB Amazon Elastic Block Store,215GB Amazon Elastic Block Store,215GB Amazon Elastic Block Store,215GB Amazon Elastic Block Store,322GB Amazon Elastic Block Store Network,,Amazon Elastic,Amazon Elastic,Amazon Elastic,Amazon Elastic,Amazon Elastic OS,,Ubuntu 22.04,Ubuntu 22.04,Ubuntu 22.04,Ubuntu 22.04,Ubuntu 22.04 Kernel,,5.19.0-1025-aws (aarch64),5.19.0-1025-aws (aarch64),5.19.0-1025-aws (aarch64),5.19.0-1025-aws (aarch64),5.19.0-1025-aws (x86_64) Compiler,,GCC 11.3.0,GCC 11.3.0,GCC 11.3.0,GCC 11.3.0,GCC 11.4.0 File-System,,ext4,ext4,ext4,ext4,ext4 System Layer,,amazon,amazon,amazon,amazon,amazon Vulkan,,,,,,1.3.238 ,,"m7g.16xlarge Graviton3","c6g.16xlarge Graviton2","c7g.16xlarge Graviton3","c7gn.16xlarge Graviton3E","c6a.16xlarge AMD Zen 3" "NAS Parallel Benchmarks - Test / Class: LU.C (Mop/s)",HIB,28341.68,18741.90,28375.71,28369.11,95221.40 "Liquid-DSP - Threads: 32 - Buffer Length: 256 - Filter Length: 512 (samples/s)",HIB,81396667,67486333,81412000,81394000,274803333 "OpenSSL - Algorithm: RSA4096 (sign/s)",HIB,10181.9,2624.3,10181.4,10183.3,8392.4 "srsRAN Project - Test: Downlink Processor Benchmark (Mbps)",HIB,318.5,197.2,319.7,323.2,691.3 "NAS Parallel Benchmarks - Test / Class: SP.C (Mop/s)",HIB,17244.85,9711.70,17219.95,17163.11,34025.35 "Liquid-DSP - Threads: 64 - Buffer Length: 256 - Filter Length: 512 (samples/s)",HIB,162753333,134926667,162766667,162756667,460076667 "srsRAN Project - Test: PUSCH Processor Benchmark, Throughput Thread (Mbps)",HIB,95.8,63.8,95.7,97.4,215.9 "OpenSSL - Algorithm: RSA4096 (verify/s)",HIB,713859.5,214040.9,713945.9,713754.8,548396.5 "Liquid-DSP - Threads: 32 - Buffer Length: 256 - Filter Length: 57 (samples/s)",HIB,721493333,489270000,721386667,721380000,1444266667 "Graph500 - Scale: 26 (bfs max_TEPS)",HIB,1227790000,874389000,1206990000,1207760000,417777000 "Graph500 - Scale: 26 (bfs median_TEPS)",HIB,1194320000,860432000,1177710000,1175640000,410571000 "OpenSSL - Algorithm: AES-256-GCM (byte/s)",HIB,283333113630,129199593157,283373795737,351152465420,138457889450 "OpenSSL - Algorithm: AES-128-GCM (byte/s)",HIB,332033171900,158436163857,332064349843,411130469943,151449269317 "ACES DGEMM - Sustained Floating-Point Rate (GFLOP/s)",HIB,24.362353,20.417952,24.140605,24.078529,9.388050 "Stress-NG - Test: Memory Copying (Bogo Ops/s)",HIB,20484.24,11324.79,20478.67,20475.96,8080.43 "Stress-NG - Test: Matrix Math (Bogo Ops/s)",HIB,368750.67,284713.63,368671.39,369258.89,147576.41 "Stress-NG - Test: Vector Shuffle (Bogo Ops/s)",HIB,54143.40,35614.51,54472.07,54695.04,22255.84 "nekRS - Input: Kershaw (flops/rank)",HIB,3150680000,1760336667,3261853333,3302823333,4308810000 "Stress-NG - Test: Matrix 3D Math (Bogo Ops/s)",HIB,10403.93,5752.17,10813.59,10882.02,4571.96 "Monte Carlo Simulations of Ionised Nebulae - Input: Dust 2D tau100.0 (sec)",LIB,82.669,145.374,82.822,82.974,194.435 "Xcompact3d Incompact3d - Input: input.i3d 129 Cells Per Direction (sec)",LIB,3.09871038,5.63720735,3.14447999,3.11489828,7.01975288 "Stress-NG - Test: Vector Floating Point (Bogo Ops/s)",HIB,76102.55,42850.82,76178.46,76911.74,96529.51 "OpenSSL - Algorithm: SHA512 (byte/s)",HIB,32125448870,14393925490,32145914147,32126059040,15291283297 "Xcompact3d Incompact3d - Input: input.i3d 193 Cells Per Direction (sec)",LIB,13.9454180,25.8825658,13.8326693,13.7606726,30.3145288 "Rodinia - Test: OpenMP CFD Solver (sec)",LIB,4.375,6.051,4.442,4.429,9.342 "Algebraic Multi-Grid Benchmark - (Figure Of Merit)",HIB,1646761667,1035586333,1765277667,1765966333,836999300 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512 (GFLOP/s)",HIB,88.0482,42.8284,88.1842,88.4551,44.3176 "Stress-NG - Test: Fused Multiply-Add (Bogo Ops/s)",HIB,63762252.76,37732190.54,63818458.61,63723431.55,30920910.92 "OpenSSL - Algorithm: ChaCha20 (byte/s)",HIB,103226784517,67292541203,103275516997,114118119423,138389378753 "Graph500 - Scale: 26 (sssp max_TEPS)",HIB,419754000,284689000,415758000,411762000,204550000 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512 (GFLOP/s)",HIB,84.4739,44.9297,84.7451,85.0060,42.4394 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512 (GFLOP/s)",HIB,162.956,81.9412,163.276,163.559,82.7584 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256 (GFLOP/s)",HIB,40.8923,20.6279,40.8283,40.9708,20.8719 "OpenSSL - Algorithm: ChaCha20-Poly1305 (byte/s)",HIB,74287460990,46717636807,74318842213,79969465487,92522999373 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512 (GFLOP/s)",HIB,46.2504,24.2658,46.3706,46.5300,23.5212 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256 (GFLOP/s)",HIB,78.5049,40.1104,77.7685,78.1658,41.5868 "nekRS - Input: TurboPipe Periodic (flops/rank)",HIB,3976300000,2220190000,3978983333,4141440000,4337536667 "NAS Parallel Benchmarks - Test / Class: MG.C (Mop/s)",HIB,50126.29,25671.29,49742.30,49860.68,45946.81 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: float - X Y Z: 256 (GFLOP/s)",HIB,81.4442,41.9816,81.0096,81.1671,43.5907 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128 (GFLOP/s)",HIB,306.540,209.496,301.418,300.396,158.858 "LAMMPS Molecular Dynamics Simulator - Model: Rhodopsin Protein (ns/day)",HIB,37.558,25.950,37.412,37.482,19.563 "Graph500 - Scale: 26 (sssp median_TEPS)",HIB,299497000,209350000,293826000,296164000,157688000 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128 (GFLOP/s)",HIB,186.356,135.358,184.026,184.110,98.7026 "LAMMPS Molecular Dynamics Simulator - Model: 20k Atoms (ns/day)",HIB,36.927,25.171,36.862,36.838,20.342 "Pennant - Test: leblancbig (Hydro Cycle Time - sec)",LIB,6.720537,12.17683,6.961345,6.839998,9.917565 "NWChem - Input: C240 Buckyball (sec)",LIB,1940.2,2976.9,1962.7,1914,3440.4 "Pennant - Test: sedovbig (Hydro Cycle Time - sec)",LIB,9.206490,16.48050,9.422270,9.340953,16.53050 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: float - X Y Z: 256 (GFLOP/s)",HIB,164.873,92.3996,162.010,162.361,102.652 "Liquid-DSP - Threads: 64 - Buffer Length: 256 - Filter Length: 57 (samples/s)",HIB,1442400000,978200000,1442366667,1442666667,1710800000 "GROMACS - Implementation: MPI CPU - Input: water_GMX50_bare (Ns/Day)",HIB,4.223,2.767,4.200,4.820,3.965 "LULESH - (z/s)",HIB,28296.378,17557.485,28708.656,28736.226,16708.258 "nginx - Connections: 500 (Reqs/sec)",HIB,255768.44,148964.69,255145.52,253518.51,165847.75 "HeFFTe - Highly Efficient FFT for Exascale - Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128 (GFLOP/s)",HIB,138.014,81.4498,133.514,133.422,86.3730 "NAS Parallel Benchmarks - Test / Class: CG.C (Mop/s)",HIB,21988.99,13103.62,21911.02,22155.36,20210.00 "NAS Parallel Benchmarks - Test / Class: EP.D (Mop/s)",HIB,3738.98,2216.26,3664.54,3657.67,3061.42 "QMCPACK - Input: simple-H2O (Execution Time - sec)",LIB,28.041,45.225,27.990,27.999,26.867 "srsRAN Project - Test: PUSCH Processor Benchmark, Throughput Total (Mbps)",HIB,5413.8,3938.7,5356.8,5431.2,6479.1 "GPAW - Input: Carbon Nanotube (sec)",LIB,61.831,92.760,62.083,56.440,89.818 "QMCPACK - Input: FeCO6_b3lyp_gms (Execution Time - sec)",LIB,211.60,302.19,211.32,188.28,184.10 "Monte Carlo Simulations of Ionised Nebulae - Input: Gas HII40 (sec)",LIB,13.575,20.758,13.659,13.525,12.669 "BRL-CAD - VGR Performance Metric (VGR Performance Metric)",HIB,783777,533020,789066,744743,485038 "LeelaChessZero - Backend: Eigen (Nodes/s)",HIB,1398,891,1382,1444,1152 "nginx - Connections: 1000 (Reqs/sec)",HIB,255616.04,158676.40,255552.05,256585.83,163178.67 "Kripke - (Throughput FoM)",HIB,339000400,220120233,354442733,354234067,237087650 "QMCPACK - Input: FeCO6_b3lyp_gms (Execution Time - sec)",LIB,205.72,297.94,204.77,204.25,187.32 "Remhos - Test: Sample Remap Example (sec)",LIB,14.040,20.740,14.120,14.082,22.104 "Liquid-DSP - Threads: 32 - Buffer Length: 256 - Filter Length: 32 (samples/s)",HIB,1136066667,765466667,1136133333,1136000000,1193966667 "Stress-NG - Test: Wide Vector Math (Bogo Ops/s)",HIB,1542834.94,997272.65,1535336.57,1530043.52,1380146.63 "Laghos - Test: Sedov Blast Wave, ube_922_hex.mesh (Major Kernels Rate)",HIB,410.55,322.37,408.01,423.11,275.92 "Stress-NG - Test: Vector Math (Bogo Ops/s)",HIB,217235.59,147886.14,217446.12,217567.10,221776.15 "Liquid-DSP - Threads: 64 - Buffer Length: 256 - Filter Length: 32 (samples/s)",HIB,2270500000,1531400000,2271966667,2266833333,2184866667 "Timed Godot Game Engine Compilation - Time To Compile (sec)",LIB,154.378,218.276,156.687,155.951,147.737 "LeelaChessZero - Backend: BLAS (Nodes/s)",HIB,1301,947,1333,1392,1316 "QMCPACK - Input: Li2_STO_ae (Execution Time - sec)",LIB,112.61,165.12,112.64,113.20,123.95 "Rodinia - Test: OpenMP LavaMD (sec)",LIB,43.788,62.224,43.963,44.044,64.179 "7-Zip Compression - Test: Compression Rating (MIPS)",HIB,316825,240702,311056,312009,230970 "Laghos - Test: Triple Point Problem (Major Kernels Rate)",HIB,232.01,180.80,230.68,236.22,227.40 "Coremark - CoreMark Size 666 - Iterations Per Second (Iterations/Sec)",HIB,1601880.342264,1260642.177024,1605948.674645,1611801.559265,1466587.036580 "OpenSSL - Algorithm: SHA256 (byte/s)",HIB,54212515580,42472798847,54216561263,54154218593,45857534777 "Timed Gem5 Compilation - Time To Compile (sec)",LIB,180.247,225.305,181.779,182.471,192.118 "Timed Node.js Compilation - Time To Compile (sec)",LIB,237.783,287.814,238.543,238.636,230.423 "7-Zip Compression - Test: Decompression Rating (MIPS)",HIB,285540,234202,285633,285677,235787 "Stress-NG - Test: CPU Cache (Bogo Ops/s)",HIB,3892396.34,1921785.20,3844101.98,3860335.38,1447265.35 "Stress-NG - Test: NUMA (Bogo Ops/s)",HIB,3759.10,2112.66,3523.58,3525.17,552.68 "Stockfish - Total Time (Nodes/s)",HIB,112119711,86609284,117316476,117027121,96905609 "HeFFTe - Highly Efficient FFT for Exascale - Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128 (GFLOP/s)",HIB,57.1503,32.7468,55.1055,55.1038,48.9432 "Rodinia - Test: OpenMP Streamcluster (sec)",LIB,11.663,13.735,11.625,10.690,8.396