Gigabyte G242-P36 Ampere Altra Max Server

Benchmarks by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2401176-NE-GIGABYTEG67&grr&sro.

PyTorch

Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-152

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-50

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-152

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-50

Xmrig

Variant: Wownero - Hash Count: 1M

Speedb

Test: Sequential Fill

Xmrig

Variant: Monero - Hash Count: 1M

Neural Magic DeepSparse

Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Timed LLVM Compilation

Build System: Unix Makefiles

LeelaChessZero

Backend: BLAS

LeelaChessZero

Backend: Eigen

Neural Magic DeepSparse

Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream

Quicksilver

Input: CTS2

Timed Linux Kernel Compilation

Build: allmodconfig

Stress-NG

Test: Atomic

Timed LLVM Compilation

Build System: Ninja

Llama.cpp

Model: llama-2-70b-chat.Q5_0.gguf

Stockfish

Total Time

OpenSSL

Algorithm: ChaCha20-Poly1305

OpenSSL

Algorithm: ChaCha20

OpenSSL

Algorithm: AES-256-GCM

Speedb

Test: Read While Writing

RocksDB

Test: Random Read

Quicksilver

Input: CORAL2 P2

OpenSSL

Algorithm: AES-128-GCM

OpenSSL

Algorithm: SHA256

OpenSSL

Algorithm: SHA512

Speedb

Test: Random Read

RocksDB

Test: Read While Writing

Llama.cpp

Model: llama-2-13b.Q4_0.gguf

CacheBench

Test: Read

CacheBench

Test: Read / Modify / Write

CacheBench

Test: Write

RocksDB

Test: Read Random Write Random

Stress-NG

Test: Futex

Stress-NG

Test: Context Switching

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream

Algebraic Multi-Grid Benchmark

Timed Linux Kernel Compilation

Build: defconfig

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

Stress-NG

Test: MEMFD

Speedb

Test: Random Fill

Speedb

Test: Random Fill Sync

Speedb

Test: Update Random

Speedb

Test: Read Random Write Random

RocksDB

Test: Update Random

OpenSSL

Algorithm: RSA4096

OpenSSL

Algorithm: RSA4096

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream

Quicksilver

Input: CORAL2 P1

Neural Magic DeepSparse

Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream

Neural Magic DeepSparse

Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream

7-Zip Compression

Test: Decompression Rating

7-Zip Compression

Test: Compression Rating

Stress-NG

Test: IO_uring

Stress-NG

Test: MMAP

Stress-NG

Test: Cloning

Stress-NG

Test: Malloc

Stress-NG

Test: CPU Cache

Stress-NG

Test: Pthread

Stress-NG

Test: Zlib

Stress-NG

Test: Vector Shuffle

Stress-NG

Test: Vector Math

Stress-NG

Test: Wide Vector Math

Stress-NG

Test: Matrix Math

Stress-NG

Test: Function Call

Stress-NG

Test: Matrix 3D Math

Stress-NG

Test: CPU Stress

Stress-NG

Test: AVL Tree

Stress-NG

Test: Crypto

Stress-NG

Test: Fused Multiply-Add

Stress-NG

Test: Hash

Stress-NG

Test: SENDFILE

Stress-NG

Test: AVX-512 VNNI

Stress-NG

Test: Glibc Qsort Data Sorting

Stress-NG

Test: Vector Floating Point

Stress-NG

Test: Floating Point

Stress-NG

Test: Poll

Stress-NG

Test: Glibc C String Functions

Stress-NG

Test: System V Message Passing

Stress-NG

Test: Forking

Stress-NG

Test: Memory Copying

Stress-NG

Test: Semaphores

Stress-NG

Test: Mutex

Stress-NG

Test: Mixed Scheduler

Stress-NG

Test: NUMA

Stress-NG

Test: Pipe

Stress-NG

Test: Socket Activity

Llama.cpp

Model: llama-2-7b.Q4_0.gguf

miniFE

Problem Size: Small

ACES DGEMM

Sustained Floating-Point Rate

Phoronix Test Suite v10.8.5