AWS Graviton1 To Graviton4 Benchmarks

Benchmarks by Michael Larabel for a future article looking at AWS Graviton1, Graviton2, Graviton3, Graviton4 metal performance.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2407243-NE-GRAVITON434
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
Show Result Confidence Charts

Limit displaying results to tests within:

Timed Code Compilation 3 Tests
C/C++ Compiler Tests 8 Tests
Compression Tests 2 Tests
CPU Massive 11 Tests
Creator Workloads 5 Tests
Database Test Suite 2 Tests
HPC - High Performance Computing 5 Tests
Imaging 3 Tests
Common Kernel Benchmarks 3 Tests
Molecular Dynamics 2 Tests
MPI Benchmarks 3 Tests
Multi-Core 14 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 5 Tests
Python 2 Tests
Scientific Computing 3 Tests
Server 4 Tests
Server CPU Tests 8 Tests
Single-Threaded 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
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per RAM Channel 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
Graviton1 16 Cores a1.metal
July 22
  18 Hours, 2 Minutes
Graviton2 64 Cores r6g.metal
July 23
  7 Hours, 20 Minutes
Graviton3 64 Cores r7g.metal
July 24
  4 Hours, 57 Minutes
Graviton4 96 Cores r8g.metal-24xl
July 23
  4 Hours, 30 Minutes
Invert Hiding All Results Option
  8 Hours, 42 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):

Graviton1 16 Cores a1.metal

Processor: ARMv8 Cortex-A72 (16 Cores), Motherboard: Amazon EC2 a1.metal (1.0 BIOS), Memory: 32GB, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-1009-aws (aarch64), Compiler: GCC 13.2.0, File-System: ext4

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Java Notes: OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Branch predictor hardening BHB + srbds: Not affected + tsx_async_abort: Not affected

Graviton4 96 Cores r8g.metal-24xl

Changed Processor to ARMv8 Neoverse-V2 (96 Cores).

Changed Motherboard to Amazon EC2 r8g.metal-24xl (1.0 BIOS).

Changed Memory to 12 x 64GB DDR5-5600MT/s.

Security Change: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Graviton2 64 Cores r6g.metal

Changed Processor to ARMv8 Neoverse-N1 (64 Cores).

Changed Motherboard to Amazon EC2 r6g.metal (1.0 BIOS).

Changed Memory to 8 x 64GB DDR4-3200MT/s.

Security Change: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Graviton3 64 Cores r7g.metal

Changed Processor to ARMv8 Neoverse-V1 (64 Cores).

Changed Motherboard to Amazon EC2 r7g.metal (1.0 BIOS).

Changed Memory to 8 x 64GB DDR5-4400MT/s.

AWS Graviton1 To Graviton4 BenchmarksProcessorMotherboardMemoryDiskNetworkOSKernelCompilerFile-SystemGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metalARMv8 Cortex-A72 (16 Cores)Amazon EC2 a1.metal (1.0 BIOS)32GB429GB Amazon Elastic Block StoreAmazon ElasticUbuntu 24.046.8.0-1009-aws (aarch64)GCC 13.2.0ext4ARMv8 Neoverse-V2 (96 Cores)Amazon EC2 r8g.metal-24xl (1.0 BIOS)12 x 64GB DDR5-5600MT/sARMv8 Neoverse-N1 (64 Cores)Amazon EC2 r6g.metal (1.0 BIOS)8 x 64GB DDR4-3200MT/sARMv8 Neoverse-V1 (64 Cores)Amazon EC2 r7g.metal (1.0 BIOS)8 x 64GB DDR5-4400MT/sOpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v Java Details- OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)Python Details- Python 3.12.3Security Details- Graviton1 16 Cores a1.metal: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Branch predictor hardening BHB + srbds: Not affected + tsx_async_abort: Not affected - Graviton4 96 Cores r8g.metal-24xl: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Graviton2 64 Cores r6g.metal: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected - Graviton3 64 Cores r7g.metal: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metalLogarithmic Result OverviewPhoronix Test SuiteSpeedbGraphicsMagickACES DGEMMsrsRAN ProjectTimed PHP Compilationlibavif avifencStockfishNumpy BenchmarkPyPerformancePHPBenchC-RayPostgreSQLGROMACSLZ4 CompressionWebP Image EncodeStress-NGminiFEBlenderOpenSSLTimed Node.js CompilationHigh Performance Conjugate GradientCoremark7-Zip CompressionTimed Godot Game Engine CompilationRocksDBJohn The Ripper

AWS Graviton1 To Graviton4 Benchmarkscompress-lz4: 9 - Compression Speedcompress-lz4: 9 - Decompression Speedcompress-lz4: 12 - Compression Speedcompress-lz4: 12 - Decompression Speedstress-ng: CPU Cachestress-ng: Power Mathstress-ng: Matrix Mathstress-ng: Vector Mathstress-ng: Floating Pointstress-ng: Matrix 3D Mathstress-ng: Memory Copyingstress-ng: Logarithmic Mathstress-ng: Fused Multiply-Addstress-ng: Trigonometric Mathstress-ng: Vector Floating Pointwebp: Defaultwebp: Quality 100webp: Quality 100, Losslesswebp: Quality 100, Highest Compressionnumpy: gromacs: MPI CPU - water_GMX50_bareopenssl: SHA256hpcg: 104 104 104 - 60mt-dgemm: Sustained Floating-Point Rateminife: Smallcoremark: CoreMark Size 666 - Iterations Per Secondstockfish: Chess Benchmarkcompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingjohn-the-ripper: bcryptjohn-the-ripper: WPA PSKjohn-the-ripper: Blowfishjohn-the-ripper: HMAC-SHA512john-the-ripper: MD5openssl: ChaCha20-Poly1305openssl: AES-128-GCMopenssl: AES-256-GCMopenssl: SHA512openssl: ChaCha20build-php: Time To Compilegraphics-magick: Sharpengraphics-magick: Enhancedgraphics-magick: Noise-Gaussiangraphics-magick: HWB Color Spacec-ray: 4K - 16c-ray: 5K - 16c-ray: 1080p - 16avifenc: 0avifenc: 2avifenc: 6, Losslessbuild-godot: Time To Compilebuild-nodejs: Time To Compilepyperformance: pathlibpyperformance: xml_etreepyperformance: json_loadspyperformance: crypto_pyaespyperformance: regex_compilepyperformance: python_startupsrsran: PDSCH Processor Benchmark, Throughput Totalsrsran: PUSCH Processor Benchmark, Throughput Totalsrsran: PDSCH Processor Benchmark, Throughput Threadsrsran: PUSCH Processor Benchmark, Throughput Threadblender: Fishy Cat - CPU-Onlyblender: BMW27 - CPU-Onlyblender: Classroom - CPU-Onlyspeedb: Rand Readspeedb: Update Randspeedb: Read While Writingspeedb: Read Rand Write Randphpbench: PHP Benchmark Suiterocksdb: Rand Readrocksdb: Update Randrocksdb: Read While Writingrocksdb: Read Rand Write Randpgbench: 100 - 1000 - Read Onlypgbench: 100 - 1000 - Read Only - Average LatencyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal13.721102.04.831090.3631797.131609.167343.9928508.49852.8838.941997.893302.684928995.122998.753466.254.633.250.401.43104.590.31361517499333.768670.8792504159.12192273.03371524930573269741244784278948003103216671904266554146770304557413372577349645725058918778285450333479.64226202681566.9461008.693140.822668.499381.11228.2291674.4694663.27611023072.122745525.91307.3169.394.018.7952.43599.391289.612328663012816724336415526212342112325282116461612686505744581018709.81639.803573.111.953595.54735225.1723418.40631796.65438910.4012455.0222886.6232525.1250252.6665585333.0445810.37119727.5711.617.821.333.22495.786.9268560484677061.619560.63583970677.82732642.26889111833791847985149466085647862818566715823833323673331106994500034448711487974001305584505320341828314879168876089.99932529420830118.92733.6784.891130.20380.2435.237126.346289.26717.658.220.665.01017.5522062.71961.7278.769.464.0733.7269.40640938471729398114909115342911829221527458547132742311560819704710128932640.34625.502200.38.022207.62004843.769242.02284337.61165655.634700.485845.8811890.8318828.2233894303.5417338.1742695.467.555.150.802.07272.072.7313897630800320.559120.41080022864.21176422.68581143949837233491233493453494997545305647126671341667466369989301573863098131268463231871412286946367238751070178.11015312811721263.489112.87416.181251.332159.5088.969266.138647.61931.710036.711017812.48295.6727.9146.639.4146.6881.54166.7018799668336354082297802781037444204165261499728335612889934210239855091.01533.233088.69.843108.33855459.6213073.07368563.70262952.327047.639831.8920452.5727799.2342419807.9326006.5276127.779.876.721.072.64400.624.2064748241389333.079934.11445836601.51535079.54621357635786299377285729507875273450774920076671453000695343256002881311923332506066466103071457747794241563073120.47225219515626638.55868.5779.836157.27595.4016.483190.349555.66623.776.628.790.113710.113395.41054.2239.653114.6062.63128.362841057756645668758376373511064110624698969810263937164354448827111536770.867OpenBenchmarking.org

LZ4 Compression

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.10Compression Level: 9 - Compression SpeedGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal918273645SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 313.7239.8025.5033.231. (CC) gcc options: -O3 -pthread

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.10Compression Level: 9 - Decompression SpeedGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal8001600240032004000SE +/- 0.06, N = 3SE +/- 0.40, N = 3SE +/- 0.31, N = 3SE +/- 0.20, N = 31102.03573.12200.33088.61. (CC) gcc options: -O3 -pthread

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.10Compression Level: 12 - Compression SpeedGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34.8311.958.029.841. (CC) gcc options: -O3 -pthread

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.10Compression Level: 12 - Decompression SpeedGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal8001600240032004000SE +/- 2.61, N = 3SE +/- 1.89, N = 3SE +/- 0.29, N = 3SE +/- 0.17, N = 31090.33595.52207.63108.31. (CC) gcc options: -O3 -pthread

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: CPU CacheGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1000K2000K3000K4000K5000KSE +/- 3895.03, N = 3SE +/- 57877.80, N = 15SE +/- 23602.71, N = 15SE +/- 45276.44, N = 4631797.134735225.172004843.763855459.621. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Power MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal5K10K15K20K25KSE +/- 0.08, N = 3SE +/- 0.98, N = 3SE +/- 13.10, N = 3SE +/- 0.10, N = 31609.1623418.409242.0213073.071. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Matrix MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal140K280K420K560K700KSE +/- 74.42, N = 3SE +/- 9.55, N = 3SE +/- 7.14, N = 3SE +/- 12.06, N = 37343.99631796.65284337.61368563.701. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Vector MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal90K180K270K360K450KSE +/- 0.54, N = 3SE +/- 85.60, N = 3SE +/- 6.90, N = 3SE +/- 51.34, N = 328508.49438910.40165655.63262952.321. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Floating PointGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal3K6K9K12K15KSE +/- 1.60, N = 3SE +/- 2.60, N = 3SE +/- 0.17, N = 3SE +/- 0.20, N = 3852.8012455.024700.487047.631. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Matrix 3D MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal5K10K15K20K25KSE +/- 3.67, N = 3SE +/- 50.98, N = 3SE +/- 4.86, N = 3SE +/- 6.42, N = 3838.9422886.625845.889831.891. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Memory CopyingGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal7K14K21K28K35KSE +/- 0.09, N = 3SE +/- 13.60, N = 3SE +/- 1.30, N = 3SE +/- 1.92, N = 31997.8932525.1211890.8320452.571. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Logarithmic MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal11K22K33K44K55KSE +/- 0.28, N = 3SE +/- 1.22, N = 3SE +/- 0.91, N = 3SE +/- 0.23, N = 33302.6850252.6618828.2227799.231. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Fused Multiply-AddGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal14M28M42M56M70MSE +/- 2472.21, N = 3SE +/- 1063.93, N = 3SE +/- 1364.97, N = 3SE +/- 3002.12, N = 34928995.1265585333.0433894303.5442419807.931. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Trigonometric MathGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal10K20K30K40K50KSE +/- 0.04, N = 3SE +/- 11.48, N = 3SE +/- 2.00, N = 3SE +/- 1.81, N = 32998.7545810.3717338.1726006.521. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.17.08Test: Vector Floating PointGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal30K60K90K120K150KSE +/- 0.38, N = 3SE +/- 17.26, N = 3SE +/- 35.00, N = 3SE +/- 29.81, N = 33466.25119727.5742695.4676127.771. (CXX) g++ options: -O2 -std=gnu99 -lc -lm

WebP Image Encode

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: DefaultGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal3691215SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 34.6311.617.559.871. (CC) gcc options: -fvisibility=hidden -O2 -lm

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal246810SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 33.257.825.156.721. (CC) gcc options: -fvisibility=hidden -O2 -lm

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100, LosslessGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal0.29930.59860.89791.19721.4965SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.401.330.801.071. (CC) gcc options: -fvisibility=hidden -O2 -lm

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.4Encode Settings: Quality 100, Highest CompressionGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal0.72451.4492.17352.8983.6225SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.433.222.072.641. (CC) gcc options: -fvisibility=hidden -O2 -lm

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal110220330440550SE +/- 0.23, N = 3SE +/- 0.76, N = 3SE +/- 0.29, N = 3SE +/- 0.77, N = 3104.59495.78272.07400.62

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal246810SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 30.3136.9262.7314.2061. (CXX) g++ options: -O3 -lm

OpenSSL

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: SHA256Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20000M40000M60000M80000M100000MSE +/- 62991235.00, N = 3SE +/- 7443314.27, N = 3SE +/- 308447238.10, N = 3SE +/- 21492664.43, N = 361517499338560484677038976308003474824138931. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1428425670SE +/- 0.00496, N = 3SE +/- 0.01489, N = 3SE +/- 0.00037, N = 3SE +/- 0.01313, N = 33.7686761.6195020.5591033.079901. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1428425670SE +/- 0.003641, N = 3SE +/- 0.164173, N = 3SE +/- 0.156541, N = 3SE +/- 0.005519, N = 30.87925060.63583920.41080034.1144581. (CC) gcc options: -O3 -march=native -fopenmp

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal15K30K45K60K75KSE +/- 1.93, N = 3SE +/- 233.77, N = 3SE +/- 11.77, N = 3SE +/- 30.17, N = 34159.1270677.8022864.2036601.501. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal600K1200K1800K2400K3000KSE +/- 24.05, N = 3SE +/- 64.82, N = 3SE +/- 300.99, N = 3SE +/- 10334.27, N = 13192273.032732642.271176422.691535079.551. (CC) gcc options: -O2 -lrt" -lrt

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 1024 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal30M60M90M120M150MSE +/- 39945.87, N = 9SE +/- 2918876.82, N = 15SE +/- 1121445.57, N = 12SE +/- 1588787.71, N = 12249305711833791843949837576357861. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver

7-Zip Compression

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Compression RatingGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal100K200K300K400K500KSE +/- 28.98, N = 3SE +/- 2385.17, N = 3SE +/- 159.81, N = 3SE +/- 865.25, N = 3326974798512334912993771. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Decompression RatingGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal110K220K330K440K550KSE +/- 50.81, N = 3SE +/- 158.70, N = 3SE +/- 32.36, N = 3SE +/- 38.31, N = 3412444946602334932857291. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20K40K60K80K100KSE +/- 94.53, N = 15SE +/- 22.45, N = 3SE +/- 0.67, N = 3SE +/- 1.33, N = 378428564745349507871. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20K40K60K80K100KSE +/- 1.33, N = 3SE +/- 0.00, N = 3SE +/- 11.59, N = 3SE +/- 6.00, N = 378948628149975527341. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20K40K60K80K100KSE +/- 106.90, N = 15SE +/- 3.00, N = 3SE +/- 36.69, N = 3SE +/- 13.72, N = 380038566745305507741. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: HMAC-SHA512Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal30M60M90M120M150MSE +/- 9769.57, N = 3SE +/- 162829.50, N = 3SE +/- 77846.72, N = 3SE +/- 68676.37, N = 31032166715823833364712667920076671. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal500K1000K1500K2000K2500KSE +/- 287.69, N = 3SE +/- 1201.85, N = 3SE +/- 333.33, N = 3SE +/- 1154.70, N = 31904262367333134166714530001. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenSSL

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20-Poly1305Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20000M40000M60000M80000M100000MSE +/- 4948671.03, N = 3SE +/- 512363.59, N = 3SE +/- 281207.00, N = 3SE +/- 7386579.63, N = 3655414677011069945000346636998930695343256001. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-128-GCMGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal100000M200000M300000M400000M500000MSE +/- 35254075.37, N = 3SE +/- 2173107.96, N = 3SE +/- 27172495.86, N = 3SE +/- 2741312.53, N = 3304557413374448711487971573863098132881311923331. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-256-GCMGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal90000M180000M270000M360000M450000MSE +/- 17186337.18, N = 3SE +/- 15611507.33, N = 3SE +/- 5879487.16, N = 3SE +/- 48618733.94, N = 3257734964574001305584501268463231872506066466101. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: SHA512Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal11000M22000M33000M44000M55000MSE +/- 4619831.10, N = 3SE +/- 4634061.90, N = 3SE +/- 20829920.25, N = 3SE +/- 2215754.50, N = 325058918775320341828314122869463307145774771. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal30000M60000M90000M120000M150000MSE +/- 4780384.74, N = 3SE +/- 430328.22, N = 3SE +/- 157650.92, N = 3SE +/- 22981194.52, N = 3828545033314879168876067238751070942415630731. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

Timed PHP Compilation

This test times how long it takes to build PHP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.3.4Time To CompileGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal100200300400500SE +/- 0.12, N = 3SE +/- 0.10, N = 3SE +/- 0.59, N = 3SE +/- 0.09, N = 3479.6490.00178.11120.47

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SharpenGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal70140210280350SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3263251532521. (CC) gcc options: -fopenmp -O2 -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal60120180240300SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3202941281951. (CC) gcc options: -fopenmp -O2 -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal50100150200250SE +/- 0.00, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3262081171561. (CC) gcc options: -fopenmp -O2 -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: HWB Color SpaceGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal70140210280350SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3813012122661. (CC) gcc options: -fopenmp -O2 -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread -lgomp

C-Ray

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 4K - Rays Per Pixel: 16Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal120240360480600SE +/- 1.65, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3566.9518.9363.4938.561. (CC) gcc options: -lpthread -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 5K - Rays Per Pixel: 16Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal2004006008001000SE +/- 2.98, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 31008.6933.68112.8768.581. (CC) gcc options: -lpthread -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 1080p - Rays Per Pixel: 16Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal306090120150SE +/- 0.248, N = 3SE +/- 0.007, N = 3SE +/- 0.010, N = 3SE +/- 0.010, N = 3140.8224.89116.1819.8361. (CC) gcc options: -lpthread -lm

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 0Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal140280420560700SE +/- 3.02, N = 3SE +/- 0.01, N = 3SE +/- 0.17, N = 3SE +/- 0.07, N = 3668.50130.20251.33157.281. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 2Graviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal80160240320400SE +/- 0.85, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3381.1180.24159.5195.401. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 1.0Encoder Speed: 6, LosslessGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal714212835SE +/- 0.098, N = 3SE +/- 0.046, N = 3SE +/- 0.034, N = 3SE +/- 0.086, N = 328.2295.2378.9696.4831. (CXX) g++ options: -O3 -fPIC -lm

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal400800120016002000SE +/- 0.31, N = 3SE +/- 1.14, N = 3SE +/- 0.92, N = 3SE +/- 0.44, N = 31674.47126.35266.14190.35

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To CompileGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal10002000300040005000SE +/- 1.44, N = 3SE +/- 0.69, N = 3SE +/- 0.21, N = 3SE +/- 0.73, N = 34663.28289.27647.62555.67

PyPerformance

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: pathlibGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal20406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3110.017.631.723.7

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: xml_etreeGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal50100150200250SE +/- 0.58, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3230.058.2100.076.6

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: json_loadsGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1632486480SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 372.120.636.728.7

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: crypto_pyaesGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal50100150200250SE +/- 0.33, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3227.065.0110.090.1

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: regex_compileGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal100200300400500SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3455101178137

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.11Benchmark: python_startupGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 325.907.5512.4010.10

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput TotalGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal5K10K15K20K25KSE +/- 10.32, N = 3SE +/- 15.31, N = 3SE +/- 88.68, N = 5SE +/- 189.99, N = 31307.322062.78295.613395.41. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput TotalGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal400800120016002000SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3169.31961.7727.91054.2MIN: 1132.6 / MAX: 1961.8MIN: 416MIN: 599.61. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput ThreadGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal60120180240300SE +/- 0.21, N = 3SE +/- 3.00, N = 3SE +/- 0.07, N = 3SE +/- 0.15, N = 394.0278.7146.6239.61. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput ThreadGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1530456075SE +/- 0.23, N = 4SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 318.769.439.453.0MIN: 11.3 / MAX: 18.9MIN: 45.4MIN: 24.4MIN: 35.31. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing is supported. This system/blender test profile makes use of the system-supplied Blender. Use pts/blender if wishing to stick to a fixed version of Blender. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Fishy Cat - Compute: CPU-OnlyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal2004006008001000SE +/- 0.38, N = 3SE +/- 0.26, N = 3SE +/- 0.35, N = 3SE +/- 0.29, N = 3952.4364.07146.68114.60

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: BMW27 - Compute: CPU-OnlyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal130260390520650SE +/- 0.63, N = 3SE +/- 0.03, N = 3SE +/- 0.21, N = 3SE +/- 0.18, N = 3599.3933.7281.5462.63

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Classroom - Compute: CPU-OnlyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal30060090012001500SE +/- 0.28, N = 3SE +/- 0.03, N = 3SE +/- 0.35, N = 3SE +/- 0.09, N = 31289.6169.40166.70128.36

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal140M280M420M560M700MSE +/- 174965.25, N = 15SE +/- 229391.69, N = 3SE +/- 445171.45, N = 3SE +/- 1029086.00, N = 3232866306409384711879966832841057751. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Update RandomGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal160K320K480K640K800KSE +/- 302.33, N = 3SE +/- 3844.86, N = 3SE +/- 3440.64, N = 3SE +/- 3804.56, N = 31281677293983635406645661. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read While WritingGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal2M4M6M8M10MSE +/- 20302.77, N = 8SE +/- 87574.70, N = 15SE +/- 74542.67, N = 15SE +/- 23497.61, N = 3243364111490911822978087583761. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read Random Write RandomGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1.1M2.2M3.3M4.4M5.5MSE +/- 214.78, N = 3SE +/- 4241.50, N = 3SE +/- 25580.42, N = 3SE +/- 3662.12, N = 35526215342911278103737351101. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal200K400K600K800K1000KSE +/- 481.21, N = 3SE +/- 273.39, N = 3SE +/- 2661.11, N = 3SE +/- 964.81, N = 3234211829221444204641106

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal110M220M330M440M550MSE +/- 12274.71, N = 3SE +/- 53499.62, N = 3SE +/- 1103933.53, N = 15SE +/- 86390.65, N = 3232528215274585471652614992469896981. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Update RandomGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal300K600K900K1200K1500KSE +/- 443.99, N = 3SE +/- 5313.65, N = 3SE +/- 4819.65, N = 15SE +/- 13144.46, N = 3164616132742372833510263931. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal2M4M6M8M10MSE +/- 16876.08, N = 3SE +/- 18723.10, N = 3SE +/- 57133.93, N = 15SE +/- 18610.37, N = 3126865011560819612889971643541. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read Random Write RandomGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal1.5M3M4.5M6M7.5MSE +/- 4162.42, N = 3SE +/- 15765.85, N = 3SE +/- 5811.82, N = 3SE +/- 18849.94, N = 35744587047101342102344882711. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal600K1200K1800K2400K3000KSE +/- 107.42, N = 3SE +/- 15759.90, N = 3SE +/- 7623.84, N = 10SE +/- 3990.43, N = 3101870289326498550911536771. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average LatencyGraviton1 16 Cores a1.metalGraviton4 96 Cores r8g.metal-24xlGraviton2 64 Cores r6g.metalGraviton3 64 Cores r7g.metal3691215SE +/- 0.010, N = 3SE +/- 0.002, N = 3SE +/- 0.008, N = 10SE +/- 0.003, N = 39.8160.3461.0150.8671. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

76 Results Shown

LZ4 Compression:
  9 - Compression Speed
  9 - Decompression Speed
  12 - Compression Speed
  12 - Decompression Speed
Stress-NG:
  CPU Cache
  Power Math
  Matrix Math
  Vector Math
  Floating Point
  Matrix 3D Math
  Memory Copying
  Logarithmic Math
  Fused Multiply-Add
  Trigonometric Math
  Vector Floating Point
WebP Image Encode:
  Default
  Quality 100
  Quality 100, Lossless
  Quality 100, Highest Compression
Numpy Benchmark
GROMACS
OpenSSL
High Performance Conjugate Gradient
ACES DGEMM
miniFE
Coremark
Stockfish
7-Zip Compression:
  Compression Rating
  Decompression Rating
John The Ripper:
  bcrypt
  WPA PSK
  Blowfish
  HMAC-SHA512
  MD5
OpenSSL:
  ChaCha20-Poly1305
  AES-128-GCM
  AES-256-GCM
  SHA512
  ChaCha20
Timed PHP Compilation
GraphicsMagick:
  Sharpen
  Enhanced
  Noise-Gaussian
  HWB Color Space
C-Ray:
  4K - 16
  5K - 16
  1080p - 16
libavif avifenc:
  0
  2
  6, Lossless
Timed Godot Game Engine Compilation
Timed Node.js Compilation
PyPerformance:
  pathlib
  xml_etree
  json_loads
  crypto_pyaes
  regex_compile
  python_startup
srsRAN Project:
  PDSCH Processor Benchmark, Throughput Total
  PUSCH Processor Benchmark, Throughput Total
  PDSCH Processor Benchmark, Throughput Thread
  PUSCH Processor Benchmark, Throughput Thread
Blender:
  Fishy Cat - CPU-Only
  BMW27 - CPU-Only
  Classroom - CPU-Only
Speedb:
  Rand Read
  Update Rand
  Read While Writing
  Read Rand Write Rand
PHPBench
RocksDB:
  Rand Read
  Update Rand
  Read While Writing
  Read Rand Write Rand
PostgreSQL:
  100 - 1000 - Read Only
  100 - 1000 - Read Only - Average Latency

Graviton1 16 Cores a1.metal

Processor: ARMv8 Cortex-A72 (16 Cores), Motherboard: Amazon EC2 a1.metal (1.0 BIOS), Memory: 32GB, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-1009-aws (aarch64), Compiler: GCC 13.2.0, File-System: ext4

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Java Notes: OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Branch predictor hardening BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 22 July 2024 16:03 by user ubuntu.

Graviton4 96 Cores r8g.metal-24xl

Processor: ARMv8 Neoverse-V2 (96 Cores), Motherboard: Amazon EC2 r8g.metal-24xl (1.0 BIOS), Memory: 12 x 64GB DDR5-5600MT/s, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-1009-aws (aarch64), Compiler: GCC 13.2.0, File-System: ext4

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Java Notes: OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 23 July 2024 14:13 by user ubuntu.

Graviton2 64 Cores r6g.metal

Processor: ARMv8 Neoverse-N1 (64 Cores), Motherboard: Amazon EC2 r6g.metal (1.0 BIOS), Memory: 8 x 64GB DDR4-3200MT/s, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-1009-aws (aarch64), Compiler: GCC 13.2.0, File-System: ext4

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Java Notes: OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 23 July 2024 20:03 by user ubuntu.

Graviton3 64 Cores r7g.metal

Processor: ARMv8 Neoverse-V1 (64 Cores), Motherboard: Amazon EC2 r7g.metal (1.0 BIOS), Memory: 8 x 64GB DDR5-4400MT/s, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-1009-aws (aarch64), Compiler: GCC 13.2.0, File-System: ext4

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Java Notes: OpenJDK Runtime Environment (build 11.0.23+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 24 July 2024 10:33 by user ubuntu.