Benchmarks by Michael Larabel for a future article.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2303280-NE-2303286NE24
Google Cloud c3 Sapphire Rapids vs. AMD Milan
Benchmarks by Michael Larabel for a future article.
c3-highcpu-8 SPR:
Processor: Intel Xeon Platinum 8481C (4 Cores / 8 Threads), Motherboard: Google Compute Engine c3-highcpu-8, Chipset: Intel 440FX 82441FX PMC, Memory: 16GB, Disk: 322GB nvme_card-pd, Network: Google Compute Engine Virtual
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
c2-standard-8 CLX:
Processor: Intel Xeon (4 Cores / 8 Threads), Motherboard: Google Compute Engine c2-standard-8, Chipset: Intel 440FX 82441FX PMC, Memory: 32GB, Disk: 322GB PersistentDisk, Network: Red Hat Virtio device
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
n2-standard-8 CLX:
Processor: Intel Xeon (4 Cores / 8 Threads), Motherboard: Google Compute Engine n2-standard-8, Chipset: Intel 440FX 82441FX PMC, Memory: 32GB, Disk: 322GB PersistentDisk, Network: Red Hat Virtio device
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
n2-highcpu-8 CLX:
Processor: Intel Xeon (4 Cores / 8 Threads), Motherboard: Google Compute Engine n2-highcpu-8, Chipset: Intel 440FX 82441FX PMC, Memory: 8GB, Disk: 322GB PersistentDisk, Network: Red Hat Virtio device
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
t2d-standard-8 AMD:
Processor: AMD EPYC 7B13 (8 Cores), Motherboard: Google Compute Engine t2d-standard-8, Chipset: Intel 440FX 82441FX PMC, Memory: 32GB, Disk: 322GB PersistentDisk, Network: Red Hat Virtio device
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
c2d-highcpu-8 AMD:
Processor: AMD EPYC 7B13 (4 Cores / 8 Threads), Motherboard: Google Compute Engine c2d-highcpu-8, Chipset: Intel 440FX 82441FX PMC, Memory: 16GB, Disk: 322GB PersistentDisk, Network: Red Hat Virtio device
OS: Ubuntu 22.10, Kernel: 5.19.0-1015-gcp (x86_64), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, System Layer: KVM
7-Zip Compression 22.01
Test: Compression Rating
MIPS > Higher Is Better
c3-highcpu-8 SPR ... 35306 |=====================================
c2-standard-8 CLX .. 30989 |=================================
n2-standard-8 CLX .. 28062 |==============================
n2-highcpu-8 CLX ... 28093 |==============================
t2d-standard-8 AMD . 48449 |===================================================
c2d-highcpu-8 AMD .. 39203 |=========================================
Blender 3.4
Blend File: BMW27 - Compute: CPU-Only
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 315.24 |==================================================
BRL-CAD 7.34
VGR Performance Metric
VGR Performance Metric > Higher Is Better
c3-highcpu-8 SPR ... 71072 |=====================================
c2-standard-8 CLX .. 50314 |==========================
n2-standard-8 CLX .. 44563 |=======================
n2-highcpu-8 CLX ... 44556 |=======================
t2d-standard-8 AMD . 99168 |===================================================
c2d-highcpu-8 AMD .. 67437 |===================================
CockroachDB 22.2
Workload: KV, 50% Reads - Concurrency: 128
ops/s > Higher Is Better
c3-highcpu-8 SPR ... 19321.6 |=================================================
c2-standard-8 CLX .. 13684.5 |===================================
n2-standard-8 CLX .. 12699.1 |================================
n2-highcpu-8 CLX ... 12635.1 |================================
t2d-standard-8 AMD . 19110.4 |================================================
c2d-highcpu-8 AMD .. 17168.4 |============================================
CockroachDB 22.2
Workload: KV, 95% Reads - Concurrency: 128
ops/s > Higher Is Better
c3-highcpu-8 SPR ... 24960.1 |=================================================
c2-standard-8 CLX .. 16184.2 |================================
n2-standard-8 CLX .. 14976.8 |=============================
n2-highcpu-8 CLX ... 14884.7 |=============================
t2d-standard-8 AMD . 22148.8 |===========================================
c2d-highcpu-8 AMD .. 20571.8 |========================================
Embree 4.0.1
Binary: Pathtracer ISPC - Model: Crown
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 5.8475 |==================================================
c2-standard-8 CLX .. 3.9340 |==================================
n2-standard-8 CLX .. 3.4584 |==============================
n2-highcpu-8 CLX ... 3.4714 |==============================
t2d-standard-8 AMD . 5.4928 |===============================================
c2d-highcpu-8 AMD .. 4.9231 |==========================================
Embree 4.0.1
Binary: Pathtracer ISPC - Model: Asian Dragon
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 7.3642 |==================================================
c2-standard-8 CLX .. 5.1542 |===================================
n2-standard-8 CLX .. 4.5768 |===============================
n2-highcpu-8 CLX ... 4.5994 |===============================
t2d-standard-8 AMD . 6.4173 |============================================
c2d-highcpu-8 AMD .. 5.8350 |========================================
Google Draco 1.5.6
Model: Lion
ms < Lower Is Better
c3-highcpu-8 SPR ... 6250 |=======================================
c2-standard-8 CLX .. 7437 |==============================================
n2-standard-8 CLX .. 8355 |====================================================
n2-highcpu-8 CLX ... 8391 |====================================================
t2d-standard-8 AMD . 6073 |======================================
c2d-highcpu-8 AMD .. 5617 |===================================
Google Draco 1.5.6
Model: Church Facade
ms < Lower Is Better
c3-highcpu-8 SPR ... 7573 |==============================
c2-standard-8 CLX .. 11462 |=============================================
n2-standard-8 CLX .. 12332 |=================================================
n2-highcpu-8 CLX ... 12882 |===================================================
t2d-standard-8 AMD . 8754 |===================================
c2d-highcpu-8 AMD .. 7442 |=============================
GROMACS 2023
Implementation: MPI CPU - Input: water_GMX50_bare
Ns Per Day > Higher Is Better
c3-highcpu-8 SPR ... 0.777 |====================================
c2-standard-8 CLX .. 0.579 |===========================
n2-standard-8 CLX .. 0.528 |=========================
n2-highcpu-8 CLX ... 0.529 |=========================
t2d-standard-8 AMD . 1.098 |===================================================
c2d-highcpu-8 AMD .. 0.690 |================================
Intel Open Image Denoise 1.4.0
Run: RT.hdr_alb_nrm.3840x2160
Images / Sec > Higher Is Better
c3-highcpu-8 SPR ... 0.24 |==========================================
c2-standard-8 CLX .. 0.22 |======================================
n2-standard-8 CLX .. 0.20 |===================================
n2-highcpu-8 CLX ... 0.20 |===================================
t2d-standard-8 AMD . 0.30 |====================================================
c2d-highcpu-8 AMD .. 0.16 |============================
Intel Open Image Denoise 1.4.0
Run: RTLightmap.hdr.4096x4096
Images / Sec > Higher Is Better
c3-highcpu-8 SPR ... 0.12 |==========================================
c2-standard-8 CLX .. 0.11 |======================================
n2-standard-8 CLX .. 0.10 |===================================
n2-highcpu-8 CLX ... 0.09 |===============================
t2d-standard-8 AMD . 0.15 |====================================================
c2d-highcpu-8 AMD .. 0.08 |============================
John The Ripper 2023.03.14
Test: bcrypt
Real C/S > Higher Is Better
c3-highcpu-8 SPR ... 6932 |=================================
c2-standard-8 CLX .. 6687 |================================
n2-standard-8 CLX .. 5970 |=============================
n2-highcpu-8 CLX ... 5962 |=============================
t2d-standard-8 AMD . 10585 |===================================================
c2d-highcpu-8 AMD .. 7620 |=====================================
John The Ripper 2023.03.14
Test: Blowfish
Real C/S > Higher Is Better
c3-highcpu-8 SPR ... 6930 |=================================
c2-standard-8 CLX .. 6684 |================================
n2-standard-8 CLX .. 5968 |=============================
n2-highcpu-8 CLX ... 5959 |=============================
t2d-standard-8 AMD . 10611 |===================================================
c2d-highcpu-8 AMD .. 7619 |=====================================
John The Ripper 2023.03.14
Test: MD5
Real C/S > Higher Is Better
c3-highcpu-8 SPR ... 765713 |==================================================
c2-standard-8 CLX .. 683546 |=============================================
n2-standard-8 CLX .. 599637 |=======================================
n2-highcpu-8 CLX ... 598971 |=======================================
t2d-standard-8 AMD . 738210 |================================================
c2d-highcpu-8 AMD .. 436585 |=============================
LeelaChessZero 0.28
Backend: BLAS
Nodes Per Second > Higher Is Better
c3-highcpu-8 SPR ... 1272 |====================================================
c2-standard-8 CLX .. 909 |=====================================
n2-standard-8 CLX .. 810 |=================================
n2-highcpu-8 CLX ... 770 |===============================
t2d-standard-8 AMD . 1202 |=================================================
c2d-highcpu-8 AMD .. 856 |===================================
LeelaChessZero 0.28
Backend: Eigen
Nodes Per Second > Higher Is Better
c3-highcpu-8 SPR ... 1221 |====================================================
c2-standard-8 CLX .. 902 |======================================
n2-standard-8 CLX .. 808 |==================================
n2-highcpu-8 CLX ... 801 |==================================
t2d-standard-8 AMD . 1039 |============================================
c2d-highcpu-8 AMD .. 743 |================================
MariaDB 11.0.1
Clients: 2048
Queries Per Second > Higher Is Better
c3-highcpu-8 SPR ... 332 |=====================================================
c2-standard-8 CLX .. 248 |========================================
n2-standard-8 CLX .. 220 |===================================
n2-highcpu-8 CLX ... 220 |===================================
t2d-standard-8 AMD . 301 |================================================
c2d-highcpu-8 AMD .. 305 |=================================================
MariaDB 11.0.1
Clients: 4096
Queries Per Second > Higher Is Better
c3-highcpu-8 SPR ... 317 |=====================================================
c2-standard-8 CLX .. 237 |========================================
n2-standard-8 CLX .. 207 |===================================
n2-highcpu-8 CLX ... 214 |====================================
t2d-standard-8 AMD . 280 |===============================================
c2d-highcpu-8 AMD .. 292 |=================================================
Memcached 1.6.18
Set To Get Ratio: 1:10
Ops/sec > Higher Is Better
c3-highcpu-8 SPR ... 1044947.13 |========================================
c2-standard-8 CLX .. 715723.53 |===========================
n2-standard-8 CLX .. 641713.60 |=========================
n2-highcpu-8 CLX ... 614742.62 |========================
t2d-standard-8 AMD . 1200527.42 |==============================================
c2d-highcpu-8 AMD .. 842830.77 |================================
Memcached 1.6.18
Set To Get Ratio: 1:100
Ops/sec > Higher Is Better
c3-highcpu-8 SPR ... 1030937.28 |=======================================
c2-standard-8 CLX .. 702291.93 |==========================
n2-standard-8 CLX .. 624980.66 |=======================
n2-highcpu-8 CLX ... 604363.37 |=======================
t2d-standard-8 AMD . 1224550.68 |==============================================
c2d-highcpu-8 AMD .. 835405.06 |===============================
miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM1
GFInst/s > Higher Is Better
c3-highcpu-8 SPR ... 188.65 |===============================
c2-standard-8 CLX .. 152.80 |=========================
n2-standard-8 CLX .. 133.95 |======================
n2-highcpu-8 CLX ... 133.95 |======================
t2d-standard-8 AMD . 304.05 |==================================================
c2d-highcpu-8 AMD .. 186.08 |===============================
miniBUDE 20210901
Implementation: OpenMP - Input Deck: BM1
Billion Interactions/s > Higher Is Better
c3-highcpu-8 SPR ... 7.546 |===============================
c2-standard-8 CLX .. 6.112 |=========================
n2-standard-8 CLX .. 5.358 |======================
n2-highcpu-8 CLX ... 5.358 |======================
t2d-standard-8 AMD . 12.162 |==================================================
c2d-highcpu-8 AMD .. 7.443 |===============================
NAMD 2.14
ATPase Simulation - 327,506 Atoms
days/ns < Lower Is Better
c3-highcpu-8 SPR ... 3.35779 |================================
c2-standard-8 CLX .. 4.44534 |===========================================
n2-standard-8 CLX .. 5.06706 |=================================================
n2-highcpu-8 CLX ... 5.05544 |=================================================
t2d-standard-8 AMD . 3.10822 |==============================
c2d-highcpu-8 AMD .. 4.45765 |===========================================
nekRS 22.0
Input: TurboPipe Periodic
FLOP/s > Higher Is Better
c3-highcpu-8 SPR ... 30667900000 |==================================
c2-standard-8 CLX .. 25100766667 |============================
n2-standard-8 CLX .. 22963566667 |=========================
t2d-standard-8 AMD . 40756366667 |=============================================
c2d-highcpu-8 AMD .. 27115200000 |==============================
Neural Magic DeepSparse 1.3.2
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 3.7873 |===============================
c2-standard-8 CLX .. 2.9308 |========================
n2-standard-8 CLX .. 2.6587 |======================
n2-highcpu-8 CLX ... 2.6268 |=====================
t2d-standard-8 AMD . 6.1611 |==================================================
c2d-highcpu-8 AMD .. 3.4194 |============================
Neural Magic DeepSparse 1.3.2
Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 528.02 |===================================
c2-standard-8 CLX .. 682.37 |=============================================
n2-standard-8 CLX .. 752.19 |=================================================
n2-highcpu-8 CLX ... 761.34 |==================================================
t2d-standard-8 AMD . 645.42 |==========================================
c2d-highcpu-8 AMD .. 584.56 |======================================
Neural Magic DeepSparse 1.3.2
Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 19.17 |=========================================
c2-standard-8 CLX .. 15.92 |==================================
n2-standard-8 CLX .. 15.36 |=================================
n2-highcpu-8 CLX ... 15.11 |================================
t2d-standard-8 AMD . 23.81 |===================================================
c2d-highcpu-8 AMD .. 13.55 |=============================
Neural Magic DeepSparse 1.3.2
Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 104.31 |===============================
c2-standard-8 CLX .. 125.47 |=====================================
n2-standard-8 CLX .. 130.02 |=======================================
n2-highcpu-8 CLX ... 132.22 |=======================================
t2d-standard-8 AMD . 167.67 |==================================================
c2d-highcpu-8 AMD .. 147.48 |============================================
Neural Magic DeepSparse 1.3.2
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 64.87 |================================================
c2-standard-8 CLX .. 59.99 |============================================
n2-standard-8 CLX .. 54.96 |=========================================
n2-highcpu-8 CLX ... 54.52 |========================================
t2d-standard-8 AMD . 68.86 |===================================================
c2d-highcpu-8 AMD .. 38.12 |============================
Neural Magic DeepSparse 1.3.2
Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 30.80 |===========================
c2-standard-8 CLX .. 33.29 |=============================
n2-standard-8 CLX .. 36.34 |================================
n2-highcpu-8 CLX ... 36.64 |================================
t2d-standard-8 AMD . 57.99 |===================================================
c2d-highcpu-8 AMD .. 52.42 |==============================================
Neural Magic DeepSparse 1.3.2
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 33.11 |=================================
c2-standard-8 CLX .. 29.18 |=============================
n2-standard-8 CLX .. 25.97 |==========================
n2-highcpu-8 CLX ... 25.69 |==========================
t2d-standard-8 AMD . 51.35 |===================================================
c2d-highcpu-8 AMD .. 28.49 |============================
Neural Magic DeepSparse 1.3.2
Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 60.39 |========================================
c2-standard-8 CLX .. 68.51 |=============================================
n2-standard-8 CLX .. 76.97 |==================================================
n2-highcpu-8 CLX ... 77.81 |===================================================
t2d-standard-8 AMD . 77.78 |===================================================
c2d-highcpu-8 AMD .. 70.13 |==============================================
Neural Magic DeepSparse 1.3.2
Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 6.5372 |========================================
c2-standard-8 CLX .. 6.1893 |======================================
n2-standard-8 CLX .. 5.7491 |===================================
n2-highcpu-8 CLX ... 5.6481 |===================================
t2d-standard-8 AMD . 8.1654 |==================================================
c2d-highcpu-8 AMD .. 4.5842 |============================
Neural Magic DeepSparse 1.3.2
Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 305.90 |===============================
c2-standard-8 CLX .. 323.08 |=================================
n2-standard-8 CLX .. 347.83 |====================================
n2-highcpu-8 CLX ... 354.05 |====================================
t2d-standard-8 AMD . 489.08 |==================================================
c2d-highcpu-8 AMD .. 436.25 |=============================================
Neural Magic DeepSparse 1.3.2
Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 16.22 |================================
c2-standard-8 CLX .. 13.99 |===========================
n2-standard-8 CLX .. 12.74 |=========================
n2-highcpu-8 CLX ... 12.88 |=========================
t2d-standard-8 AMD . 26.06 |===================================================
c2d-highcpu-8 AMD .. 14.35 |============================
Neural Magic DeepSparse 1.3.2
Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 123.29 |=======================================
c2-standard-8 CLX .. 142.89 |==============================================
n2-standard-8 CLX .. 156.90 |==================================================
n2-highcpu-8 CLX ... 155.23 |=================================================
t2d-standard-8 AMD . 153.16 |=================================================
c2d-highcpu-8 AMD .. 139.20 |============================================
Neural Magic DeepSparse 1.3.2
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
items/sec > Higher Is Better
c3-highcpu-8 SPR ... 3.7693 |==============================
c2-standard-8 CLX .. 2.9352 |========================
n2-standard-8 CLX .. 2.6597 |=====================
n2-highcpu-8 CLX ... 2.6570 |=====================
t2d-standard-8 AMD . 6.1956 |==================================================
c2d-highcpu-8 AMD .. 3.3909 |===========================
Neural Magic DeepSparse 1.3.2
Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
ms/batch < Lower Is Better
c3-highcpu-8 SPR ... 530.61 |===================================
c2-standard-8 CLX .. 681.34 |=============================================
n2-standard-8 CLX .. 751.94 |==================================================
n2-highcpu-8 CLX ... 752.68 |==================================================
t2d-standard-8 AMD . 642.98 |===========================================
c2d-highcpu-8 AMD .. 588.61 |=======================================
nginx 1.23.2
Connections: 100
Requests Per Second > Higher Is Better
c3-highcpu-8 SPR ... 36310.35 |================================================
c2-standard-8 CLX .. 25148.28 |=================================
n2-standard-8 CLX .. 22830.19 |==============================
n2-highcpu-8 CLX ... 22444.40 |==============================
t2d-standard-8 AMD . 35257.77 |===============================================
c2d-highcpu-8 AMD .. 25019.86 |=================================
nginx 1.23.2
Connections: 200
Requests Per Second > Higher Is Better
c3-highcpu-8 SPR ... 35602.10 |================================================
c2-standard-8 CLX .. 24695.91 |=================================
n2-standard-8 CLX .. 22570.18 |==============================
n2-highcpu-8 CLX ... 21431.42 |=============================
t2d-standard-8 AMD . 35225.04 |===============================================
c2d-highcpu-8 AMD .. 24677.03 |=================================
nginx 1.23.2
Connections: 500
Requests Per Second > Higher Is Better
c3-highcpu-8 SPR ... 34672.65 |===============================================
c2-standard-8 CLX .. 21957.17 |==============================
n2-standard-8 CLX .. 20146.10 |============================
n2-highcpu-8 CLX ... 19806.60 |===========================
t2d-standard-8 AMD . 35127.26 |================================================
c2d-highcpu-8 AMD .. 24265.31 |=================================
nginx 1.23.2
Connections: 1000
Requests Per Second > Higher Is Better
c3-highcpu-8 SPR ... 32118.58 |=============================================
c2-standard-8 CLX .. 21446.27 |==============================
n2-standard-8 CLX .. 19683.60 |============================
n2-highcpu-8 CLX ... 19418.08 |===========================
t2d-standard-8 AMD . 34079.62 |================================================
c2d-highcpu-8 AMD .. 23758.47 |=================================
nginx 1.23.2
Connections: 4000
Requests Per Second > Higher Is Better
c3-highcpu-8 SPR ... 32814.75 |===============================================
c2-standard-8 CLX .. 21594.94 |===============================
n2-standard-8 CLX .. 19947.56 |=============================
n2-highcpu-8 CLX ... 19477.31 |============================
t2d-standard-8 AMD . 33592.77 |================================================
c2d-highcpu-8 AMD .. 23747.52 |==================================
oneDNN 3.0
Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 1.50004 |===
c2-standard-8 CLX .. 19.52870 |===========================================
n2-standard-8 CLX .. 22.02520 |================================================
n2-highcpu-8 CLX ... 22.03560 |================================================
oneDNN 3.0
Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 5.34218 |==============================
c2-standard-8 CLX .. 8.02416 |==============================================
n2-standard-8 CLX .. 8.51864 |================================================
n2-highcpu-8 CLX ... 8.62637 |=================================================
oneDNN 3.0
Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 4.17707 |=====
c2-standard-8 CLX .. 34.18930 |==========================================
n2-standard-8 CLX .. 38.87170 |================================================
n2-highcpu-8 CLX ... 38.96760 |================================================
oneDNN 3.0
Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 1.47145 |=
c2-standard-8 CLX .. 52.94210 |==========================================
n2-standard-8 CLX .. 60.59250 |================================================
n2-highcpu-8 CLX ... 60.55490 |================================================
oneDNN 3.0
Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 3.54944 |====
c2-standard-8 CLX .. 35.91640 |==========================================
n2-standard-8 CLX .. 40.65420 |================================================
n2-highcpu-8 CLX ... 40.77870 |================================================
oneDNN 3.0
Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 4660.70 |===================================
c2-standard-8 CLX .. 5767.70 |===========================================
n2-standard-8 CLX .. 6078.81 |==============================================
n2-highcpu-8 CLX ... 6112.65 |==============================================
t2d-standard-8 AMD . 4004.68 |==============================
c2d-highcpu-8 AMD .. 6503.63 |=================================================
oneDNN 3.0
Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 2337.31 |==================================
c2-standard-8 CLX .. 2998.03 |============================================
n2-standard-8 CLX .. 3172.34 |===============================================
n2-highcpu-8 CLX ... 3157.38 |==============================================
t2d-standard-8 AMD . 2107.39 |===============================
c2d-highcpu-8 AMD .. 3335.70 |=================================================
oneDNN 3.0
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
c3-highcpu-8 SPR ... 0.968986 |=====
c2-standard-8 CLX .. 7.432600 |==========================================
n2-standard-8 CLX .. 8.506780 |================================================
n2-highcpu-8 CLX ... 8.445150 |================================================
OpenCV 4.7
Test: Core
ms < Lower Is Better
c3-highcpu-8 SPR ... 87372 |=====================
c2-standard-8 CLX .. 142770 |==================================
n2-standard-8 CLX .. 209277 |==================================================
n2-highcpu-8 CLX ... 208199 |==================================================
t2d-standard-8 AMD . 90443 |======================
c2d-highcpu-8 AMD .. 80771 |===================
OpenCV 4.7
Test: Graph API
ms < Lower Is Better
c3-highcpu-8 SPR ... 219931 |======================================
c2-standard-8 CLX .. 236186 |=========================================
n2-standard-8 CLX .. 267250 |==============================================
n2-highcpu-8 CLX ... 289162 |==================================================
t2d-standard-8 AMD . 210214 |====================================
c2d-highcpu-8 AMD .. 199470 |==================================
OpenCV 4.7
Test: Stitching
ms < Lower Is Better
c3-highcpu-8 SPR ... 214760 |======================================
c2-standard-8 CLX .. 250833 |=============================================
n2-standard-8 CLX .. 278493 |=================================================
n2-highcpu-8 CLX ... 281338 |==================================================
t2d-standard-8 AMD . 195823 |===================================
c2d-highcpu-8 AMD .. 187400 |=================================
OpenCV 4.7
Test: Image Processing
ms < Lower Is Better
c3-highcpu-8 SPR ... 128163 |====================================
c2-standard-8 CLX .. 147234 |=========================================
n2-standard-8 CLX .. 178704 |==================================================
n2-highcpu-8 CLX ... 172842 |================================================
t2d-standard-8 AMD . 114410 |================================
c2d-highcpu-8 AMD .. 117618 |=================================
OpenCV 4.7
Test: Object Detection
ms < Lower Is Better
c3-highcpu-8 SPR ... 38999 |==============================
c2-standard-8 CLX .. 58056 |=============================================
n2-standard-8 CLX .. 64284 |=================================================
n2-highcpu-8 CLX ... 66269 |===================================================
t2d-standard-8 AMD . 32014 |=========================
c2d-highcpu-8 AMD .. 37666 |=============================
OpenFOAM 10
Input: drivaerFastback, Small Mesh Size - Mesh Time
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 62.04 |=================================
c2-standard-8 CLX .. 85.52 |==============================================
n2-standard-8 CLX .. 95.06 |===================================================
n2-highcpu-8 CLX ... 91.51 |=================================================
t2d-standard-8 AMD . 48.75 |==========================
c2d-highcpu-8 AMD .. 63.92 |==================================
OpenFOAM 10
Input: drivaerFastback, Small Mesh Size - Execution Time
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 422.68 |=====================================
c2-standard-8 CLX .. 560.61 |=================================================
n2-standard-8 CLX .. 574.24 |==================================================
n2-highcpu-8 CLX ... 563.87 |=================================================
t2d-standard-8 AMD . 294.34 |==========================
c2d-highcpu-8 AMD .. 403.83 |===================================
OpenRadioss 2022.10.13
Model: Bumper Beam
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 303.38 |===================================
c2-standard-8 CLX .. 390.61 |=============================================
n2-standard-8 CLX .. 436.27 |==================================================
n2-highcpu-8 CLX ... 433.87 |==================================================
t2d-standard-8 AMD . 176.83 |====================
c2d-highcpu-8 AMD .. 286.02 |=================================
OpenRadioss 2022.10.13
Model: Cell Phone Drop Test
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 219.73 |==================================
c2-standard-8 CLX .. 291.49 |=============================================
n2-standard-8 CLX .. 322.22 |==================================================
n2-highcpu-8 CLX ... 320.36 |==================================================
t2d-standard-8 AMD . 135.99 |=====================
c2d-highcpu-8 AMD .. 213.88 |=================================
OpenRadioss 2022.10.13
Model: Bird Strike on Windshield
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 595.14 |=====================================
c2-standard-8 CLX .. 724.80 |=============================================
n2-standard-8 CLX .. 811.76 |==================================================
n2-highcpu-8 CLX ... 800.02 |=================================================
t2d-standard-8 AMD . 386.35 |========================
c2d-highcpu-8 AMD .. 621.88 |======================================
OpenRadioss 2022.10.13
Model: Rubber O-Ring Seal Installation
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 367.00 |===============================
c2-standard-8 CLX .. 523.45 |============================================
n2-standard-8 CLX .. 591.71 |==================================================
n2-highcpu-8 CLX ... 584.09 |=================================================
t2d-standard-8 AMD . 208.81 |==================
c2d-highcpu-8 AMD .. 364.33 |===============================
OpenSSL 3.1
Algorithm: SHA256
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 4283873987 |=============================
c2-standard-8 CLX .. 1318193530 |=========
n2-standard-8 CLX .. 1178513873 |========
n2-highcpu-8 CLX ... 1177333613 |========
t2d-standard-8 AMD . 6876127730 |==============================================
c2d-highcpu-8 AMD .. 5824179883 |=======================================
OpenSSL 3.1
Algorithm: SHA512
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 1568572920 |=======================
c2-standard-8 CLX .. 1465815227 |=====================
n2-standard-8 CLX .. 1308154027 |===================
n2-highcpu-8 CLX ... 1306223720 |===================
t2d-standard-8 AMD . 3158654817 |==============================================
c2d-highcpu-8 AMD .. 1863352067 |===========================
OpenSSL 3.1
Algorithm: RSA4096
sign/s > Higher Is Better
c3-highcpu-8 SPR ... 2062.7 |==================================================
c2-standard-8 CLX .. 1156.6 |============================
n2-standard-8 CLX .. 1028.6 |=========================
n2-highcpu-8 CLX ... 1028.8 |=========================
t2d-standard-8 AMD . 1765.0 |===========================================
c2d-highcpu-8 AMD .. 1025.7 |=========================
OpenSSL 3.1
Algorithm: ChaCha20
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 22091557637 |======================================
c2-standard-8 CLX .. 21346811813 |====================================
n2-standard-8 CLX .. 19055130217 |=================================
n2-highcpu-8 CLX ... 19065485903 |=================================
t2d-standard-8 AMD . 26325286910 |=============================================
c2d-highcpu-8 AMD .. 17266517790 |==============================
OpenSSL 3.1
Algorithm: AES-128-GCM
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 57594077823 |=============================================
c2-standard-8 CLX .. 23237312603 |==================
n2-standard-8 CLX .. 20765256453 |================
n2-highcpu-8 CLX ... 20753550177 |================
t2d-standard-8 AMD . 33948315373 |===========================
c2d-highcpu-8 AMD .. 18522423023 |==============
OpenSSL 3.1
Algorithm: AES-256-GCM
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 48008361573 |=============================================
c2-standard-8 CLX .. 16945000630 |================
n2-standard-8 CLX .. 15129238003 |==============
n2-highcpu-8 CLX ... 15133854923 |==============
t2d-standard-8 AMD . 31447472720 |=============================
c2d-highcpu-8 AMD .. 16968938653 |================
OpenSSL 3.1
Algorithm: ChaCha20-Poly1305
byte/s > Higher Is Better
c3-highcpu-8 SPR ... 15970781140 |=========================================
c2-standard-8 CLX .. 10919151843 |============================
n2-standard-8 CLX .. 9576513353 |========================
n2-highcpu-8 CLX ... 9567474300 |========================
t2d-standard-8 AMD . 17592600077 |=============================================
c2d-highcpu-8 AMD .. 11759326457 |==============================
OpenVKL 1.3.1
Benchmark: vklBenchmark ISPC
Items / Sec > Higher Is Better
c3-highcpu-8 SPR ... 98 |======================================================
c2-standard-8 CLX .. 70 |=======================================
n2-standard-8 CLX .. 63 |===================================
n2-highcpu-8 CLX ... 81 |=============================================
t2d-standard-8 AMD . 75 |=========================================
c2d-highcpu-8 AMD .. 62 |==================================
OSPRay Studio 0.11
Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
ms < Lower Is Better
c3-highcpu-8 SPR ... 20952 |==============================
c2-standard-8 CLX .. 30911 |============================================
n2-standard-8 CLX .. 35884 |===================================================
n2-highcpu-8 CLX ... 35193 |==================================================
t2d-standard-8 AMD . 25306 |====================================
c2d-highcpu-8 AMD .. 31370 |=============================================
OSPRay Studio 0.11
Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
ms < Lower Is Better
c3-highcpu-8 SPR ... 25819 |=====================
c2-standard-8 CLX .. 37892 |================================
n2-standard-8 CLX .. 61337 |===================================================
n2-highcpu-8 CLX ... 61150 |===================================================
t2d-standard-8 AMD . 30299 |=========================
c2d-highcpu-8 AMD .. 37227 |===============================
PostgreSQL 15
Scaling Factor: 100 - Clients: 800 - Mode: Read Only
TPS > Higher Is Better
c3-highcpu-8 SPR ... 311942 |================================================
c2-standard-8 CLX .. 173317 |===========================
n2-standard-8 CLX .. 158267 |========================
n2-highcpu-8 CLX ... 152093 |========================
t2d-standard-8 AMD . 323353 |==================================================
c2d-highcpu-8 AMD .. 249490 |=======================================
PostgreSQL 15
Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency
ms < Lower Is Better
c3-highcpu-8 SPR ... 2.565 |=========================
c2-standard-8 CLX .. 4.616 |=============================================
n2-standard-8 CLX .. 5.055 |=================================================
n2-highcpu-8 CLX ... 5.274 |===================================================
t2d-standard-8 AMD . 2.475 |========================
c2d-highcpu-8 AMD .. 3.206 |===============================
PostgreSQL 15
Scaling Factor: 100 - Clients: 1000 - Mode: Read Only
TPS > Higher Is Better
c3-highcpu-8 SPR ... 293725 |==============================================
c2-standard-8 CLX .. 169643 |==========================
n2-standard-8 CLX .. 155196 |========================
n2-highcpu-8 CLX ... 145790 |=======================
t2d-standard-8 AMD . 320827 |==================================================
c2d-highcpu-8 AMD .. 244198 |======================================
PostgreSQL 15
Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency
ms < Lower Is Better
c3-highcpu-8 SPR ... 3.405 |=========================
c2-standard-8 CLX .. 5.895 |============================================
n2-standard-8 CLX .. 6.450 |================================================
n2-highcpu-8 CLX ... 6.862 |===================================================
t2d-standard-8 AMD . 3.118 |=======================
c2d-highcpu-8 AMD .. 4.096 |==============================
SPECFEM3D 4.0
Model: Mount St. Helens
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 139.52 |===========================================
c2-standard-8 CLX .. 145.38 |=============================================
n2-standard-8 CLX .. 162.43 |==================================================
n2-highcpu-8 CLX ... 160.30 |=================================================
t2d-standard-8 AMD . 81.19 |=========================
c2d-highcpu-8 AMD .. 125.62 |=======================================
SPECFEM3D 4.0
Model: Layered Halfspace
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 372.09 |=============================================
c2-standard-8 CLX .. 374.76 |=============================================
n2-standard-8 CLX .. 415.18 |==================================================
n2-highcpu-8 CLX ... 416.41 |==================================================
t2d-standard-8 AMD . 223.11 |===========================
c2d-highcpu-8 AMD .. 324.85 |=======================================
SPECFEM3D 4.0
Model: Tomographic Model
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 143.94 |===========================================
c2-standard-8 CLX .. 150.92 |=============================================
n2-standard-8 CLX .. 166.38 |==================================================
n2-highcpu-8 CLX ... 167.35 |==================================================
t2d-standard-8 AMD . 81.81 |========================
c2d-highcpu-8 AMD .. 128.17 |======================================
SPECFEM3D 4.0
Model: Homogeneous Halfspace
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 179.55 |==========================================
c2-standard-8 CLX .. 190.79 |============================================
n2-standard-8 CLX .. 213.89 |==================================================
n2-highcpu-8 CLX ... 215.27 |==================================================
t2d-standard-8 AMD . 102.34 |========================
c2d-highcpu-8 AMD .. 164.00 |======================================
SPECFEM3D 4.0
Model: Water-layered Halfspace
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 321.63 |==========================================
c2-standard-8 CLX .. 347.81 |==============================================
n2-standard-8 CLX .. 380.13 |==================================================
n2-highcpu-8 CLX ... 374.19 |=================================================
t2d-standard-8 AMD . 197.52 |==========================
c2d-highcpu-8 AMD .. 287.42 |======================================
TensorFlow 2.10
Device: CPU - Batch Size: 16 - Model: ResNet-50
images/sec > Higher Is Better
c3-highcpu-8 SPR ... 14.20 |===================================================
c2-standard-8 CLX .. 13.33 |================================================
n2-standard-8 CLX .. 12.18 |============================================
n2-highcpu-8 CLX ... 12.18 |============================================
t2d-standard-8 AMD . 10.74 |=======================================
c2d-highcpu-8 AMD .. 8.66 |===============================
TensorFlow 2.10
Device: CPU - Batch Size: 32 - Model: ResNet-50
images/sec > Higher Is Better
c3-highcpu-8 SPR ... 14.93 |===================================================
c2-standard-8 CLX .. 14.11 |================================================
n2-standard-8 CLX .. 12.95 |============================================
n2-highcpu-8 CLX ... 12.99 |============================================
t2d-standard-8 AMD . 10.61 |====================================
c2d-highcpu-8 AMD .. 8.55 |=============================
TensorFlow 2.10
Device: CPU - Batch Size: 64 - Model: ResNet-50
images/sec > Higher Is Better
c3-highcpu-8 SPR ... 15.69 |===================================================
c2-standard-8 CLX .. 14.79 |================================================
n2-standard-8 CLX .. 13.74 |=============================================
n2-highcpu-8 CLX ... 13.70 |=============================================
t2d-standard-8 AMD . 10.48 |==================================
c2d-highcpu-8 AMD .. 8.46 |===========================
Timed FFmpeg Compilation 6.0
Time To Compile
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 120.44 |=======================================
c2-standard-8 CLX .. 139.04 |=============================================
n2-standard-8 CLX .. 154.90 |==================================================
n2-highcpu-8 CLX ... 154.90 |==================================================
t2d-standard-8 AMD . 80.65 |==========================
c2d-highcpu-8 AMD .. 119.64 |=======================================
Timed Linux Kernel Compilation 6.1
Build: defconfig
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 244.80 |=====================================
c2-standard-8 CLX .. 289.66 |============================================
n2-standard-8 CLX .. 322.38 |=================================================
n2-highcpu-8 CLX ... 327.76 |==================================================
t2d-standard-8 AMD . 184.68 |============================
c2d-highcpu-8 AMD .. 259.34 |========================================
uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Very Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 6.99 |=================================
c2-standard-8 CLX .. 5.69 |===========================
n2-standard-8 CLX .. 4.97 |========================
n2-highcpu-8 CLX ... 4.93 |========================
t2d-standard-8 AMD . 10.68 |===================================================
c2d-highcpu-8 AMD .. 7.90 |======================================
uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Super Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 7.48 |==================================
c2-standard-8 CLX .. 6.03 |===========================
n2-standard-8 CLX .. 5.28 |========================
n2-highcpu-8 CLX ... 5.17 |=======================
t2d-standard-8 AMD . 11.36 |===================================================
c2d-highcpu-8 AMD .. 8.14 |=====================================
uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 9.12 |==================================
c2-standard-8 CLX .. 7.45 |============================
n2-standard-8 CLX .. 6.51 |========================
n2-highcpu-8 CLX ... 6.41 |========================
t2d-standard-8 AMD . 13.67 |===================================================
c2d-highcpu-8 AMD .. 9.82 |=====================================
uvg266 0.4.1
Video Input: Bosphorus 1080p - Video Preset: Very Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 32.39 |==================================
c2-standard-8 CLX .. 26.30 |===========================
n2-standard-8 CLX .. 22.81 |========================
n2-highcpu-8 CLX ... 22.87 |========================
t2d-standard-8 AMD . 49.19 |===================================================
c2d-highcpu-8 AMD .. 36.45 |======================================
uvg266 0.4.1
Video Input: Bosphorus 1080p - Video Preset: Super Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 34.50 |==================================
c2-standard-8 CLX .. 27.72 |===========================
n2-standard-8 CLX .. 24.15 |========================
n2-highcpu-8 CLX ... 24.15 |========================
t2d-standard-8 AMD . 52.25 |===================================================
c2d-highcpu-8 AMD .. 37.79 |=====================================
uvg266 0.4.1
Video Input: Bosphorus 1080p - Video Preset: Ultra Fast
Frames Per Second > Higher Is Better
c3-highcpu-8 SPR ... 42.24 |===================================
c2-standard-8 CLX .. 34.47 |============================
n2-standard-8 CLX .. 30.05 |=========================
n2-highcpu-8 CLX ... 29.97 |=========================
t2d-standard-8 AMD . 61.85 |===================================================
c2d-highcpu-8 AMD .. 45.53 |======================================
Xcompact3d Incompact3d 2021-03-11
Input: input.i3d 129 Cells Per Direction
Seconds < Lower Is Better
c3-highcpu-8 SPR ... 32.39 |=========================================
c2-standard-8 CLX .. 37.76 |================================================
n2-standard-8 CLX .. 40.54 |===================================================
n2-highcpu-8 CLX ... 40.03 |==================================================
t2d-standard-8 AMD . 21.50 |===========================
c2d-highcpu-8 AMD .. 26.52 |=================================
Zstd Compression 1.5.4
Compression Level: 19 - Compression Speed
MB/s > Higher Is Better
c3-highcpu-8 SPR ... 10.30 |================================================
c2-standard-8 CLX .. 9.24 |===========================================
n2-standard-8 CLX .. 8.24 |======================================
n2-highcpu-8 CLX ... 8.29 |======================================
t2d-standard-8 AMD . 10.90 |===================================================
c2d-highcpu-8 AMD .. 11.00 |===================================================
Zstd Compression 1.5.4
Compression Level: 19 - Decompression Speed
MB/s > Higher Is Better
c3-highcpu-8 SPR ... 905.2 |====================================
c2-standard-8 CLX .. 701.1 |============================
n2-standard-8 CLX .. 649.5 |==========================
n2-highcpu-8 CLX ... 640.8 |==========================
t2d-standard-8 AMD . 1129.0 |=============================================
c2d-highcpu-8 AMD .. 1253.6 |==================================================
Zstd Compression 1.5.4
Compression Level: 19, Long Mode - Compression Speed
MB/s > Higher Is Better
c3-highcpu-8 SPR ... 6.50 |==================================================
c2-standard-8 CLX .. 5.86 |=============================================
n2-standard-8 CLX .. 5.13 |========================================
n2-highcpu-8 CLX ... 5.04 |=======================================
t2d-standard-8 AMD . 5.42 |==========================================
c2d-highcpu-8 AMD .. 6.71 |====================================================
Zstd Compression 1.5.4
Compression Level: 19, Long Mode - Decompression Speed
MB/s > Higher Is Better
c3-highcpu-8 SPR ... 907.2 |======================================
c2-standard-8 CLX .. 713.8 |==============================
n2-standard-8 CLX .. 660.3 |============================
n2-highcpu-8 CLX ... 650.0 |===========================
t2d-standard-8 AMD . 1075.8 |=============================================
c2d-highcpu-8 AMD .. 1183.8 |==================================================
Geometric Mean Of All Test Results
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Geometric Mean > Higher Is Better
c3-highcpu-8 SPR ... 54.76 |=============================================
c2-standard-8 CLX .. 41.96 |==================================
n2-standard-8 CLX .. 37.56 |===============================
n2-highcpu-8 CLX ... 37.36 |===============================
t2d-standard-8 AMD . 62.19 |===================================================
c2d-highcpu-8 AMD .. 47.05 |=======================================
Geometric Mean Of Timed Code Compilation Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.311 |====================================
c2-standard-8 CLX .. 1.122 |===============================
n2-standard-8 CLX .. 1.007 |============================
n2-highcpu-8 CLX ... 1.000 |============================
t2d-standard-8 AMD . 1.845 |===================================================
c2d-highcpu-8 AMD .. 1.279 |===================================
Geometric Mean Of C/C++ Compiler Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.648 |==============================================
c2-standard-8 CLX .. 1.124 |===============================
n2-standard-8 CLX .. 1.015 |============================
n2-highcpu-8 CLX ... 1.000 |============================
t2d-standard-8 AMD . 1.834 |===================================================
c2d-highcpu-8 AMD .. 1.350 |======================================
Geometric Mean Of Compression Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.318 |============================================
c2-standard-8 CLX .. 1.114 |=====================================
n2-standard-8 CLX .. 1.008 |=================================
n2-highcpu-8 CLX ... 1.000 |=================================
t2d-standard-8 AMD . 1.480 |=================================================
c2d-highcpu-8 AMD .. 1.544 |===================================================
Geometric Mean Of CPU Massive Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.570 |=================================================
c2-standard-8 CLX .. 1.120 |===================================
n2-standard-8 CLX .. 1.012 |================================
n2-highcpu-8 CLX ... 1.000 |===============================
t2d-standard-8 AMD . 1.638 |===================================================
c2d-highcpu-8 AMD .. 1.227 |======================================
Geometric Mean Of Creator Workloads Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.484 |============================================
c2-standard-8 CLX .. 1.143 |==================================
n2-standard-8 CLX .. 1.000 |==============================
n2-highcpu-8 CLX ... 1.004 |==============================
t2d-standard-8 AMD . 1.710 |===================================================
c2d-highcpu-8 AMD .. 1.291 |=======================================
Geometric Mean Of Cryptography Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.742 |=============================================
c2-standard-8 CLX .. 1.125 |=============================
n2-standard-8 CLX .. 1.001 |==========================
n2-highcpu-8 CLX ... 1.000 |==========================
t2d-standard-8 AMD . 1.955 |===================================================
c2d-highcpu-8 AMD .. 1.250 |=================================
Geometric Mean Of Database Test Suite
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.760 |==================================================
c2-standard-8 CLX .. 1.134 |================================
n2-standard-8 CLX .. 1.026 |=============================
n2-highcpu-8 CLX ... 1.000 |============================
t2d-standard-8 AMD . 1.796 |===================================================
c2d-highcpu-8 AMD .. 1.482 |==========================================
Geometric Mean Of Fortran Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.175 |===============================
c2-standard-8 CLX .. 1.102 |=============================
n2-standard-8 CLX .. 1.000 |==========================
n2-highcpu-8 CLX ... 1.004 |==========================
t2d-standard-8 AMD . 1.964 |===================================================
c2d-highcpu-8 AMD .. 1.334 |===================================
Geometric Mean Of Game Development Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.376 |===================================================
c2-standard-8 CLX .. 1.103 |=========================================
n2-standard-8 CLX .. 1.000 |=====================================
n2-highcpu-8 CLX ... 1.018 |=====================================
t2d-standard-8 AMD . 1.388 |===================================================
c2d-highcpu-8 AMD .. 1.091 |========================================
Geometric Mean Of HPC - High Performance Computing Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.367 |==============================================
c2-standard-8 CLX .. 1.114 |======================================
n2-standard-8 CLX .. 1.004 |==================================
n2-highcpu-8 CLX ... 1.000 |==================================
t2d-standard-8 AMD . 1.515 |===================================================
c2d-highcpu-8 AMD .. 1.149 |=======================================
Geometric Mean Of Common Kernel Benchmarks Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 2.041 |================================================
c2-standard-8 CLX .. 1.134 |===========================
n2-standard-8 CLX .. 1.019 |========================
n2-highcpu-8 CLX ... 1.000 |========================
t2d-standard-8 AMD . 2.152 |===================================================
c2d-highcpu-8 AMD .. 1.449 |==================================
Geometric Mean Of Machine Learning Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.343 |===================================================
c2-standard-8 CLX .. 1.120 |===========================================
n2-standard-8 CLX .. 1.010 |======================================
n2-highcpu-8 CLX ... 1.000 |======================================
t2d-standard-8 AMD . 1.312 |==================================================
c2d-highcpu-8 AMD .. 1.057 |========================================
Geometric Mean Of Molecular Dynamics Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.421 |======================================
c2-standard-8 CLX .. 1.089 |=============================
n2-standard-8 CLX .. 1.000 |===========================
n2-highcpu-8 CLX ... 1.015 |===========================
t2d-standard-8 AMD . 1.894 |===================================================
c2d-highcpu-8 AMD .. 1.369 |=====================================
Geometric Mean Of MPI Benchmarks Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.358 |===================================
c2-standard-8 CLX .. 1.085 |============================
n2-standard-8 CLX .. 1.000 |==========================
n2-highcpu-8 CLX ... 1.008 |==========================
t2d-standard-8 AMD . 1.981 |===================================================
c2d-highcpu-8 AMD .. 1.414 |====================================
Geometric Mean Of Multi-Core Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.466 |============================================
c2-standard-8 CLX .. 1.133 |==================================
n2-standard-8 CLX .. 1.003 |==============================
n2-highcpu-8 CLX ... 1.000 |==============================
t2d-standard-8 AMD . 1.700 |===================================================
c2d-highcpu-8 AMD .. 1.316 |=======================================
Geometric Mean Of NVIDIA GPU Compute Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.546 |=================================================
c2-standard-8 CLX .. 1.133 |====================================
n2-standard-8 CLX .. 1.019 |================================
n2-highcpu-8 CLX ... 1.000 |================================
t2d-standard-8 AMD . 1.614 |===================================================
c2d-highcpu-8 AMD .. 1.104 |===================================
Geometric Mean Of Intel oneAPI Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.523 |===================================================
c2-standard-8 CLX .. 1.157 |=======================================
n2-standard-8 CLX .. 1.000 |=================================
n2-highcpu-8 CLX ... 1.020 |==================================
t2d-standard-8 AMD . 1.510 |===================================================
c2d-highcpu-8 AMD .. 1.078 |====================================
Geometric Mean Of OpenMPI Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.324 |================================
c2-standard-8 CLX .. 1.102 |===========================
n2-standard-8 CLX .. 1.000 |========================
n2-highcpu-8 CLX ... 1.010 |========================
t2d-standard-8 AMD . 2.107 |===================================================
c2d-highcpu-8 AMD .. 1.397 |==================================
Geometric Mean Of Programmer / Developer System Benchmarks Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.326 |============================================
c2-standard-8 CLX .. 1.119 |=====================================
n2-standard-8 CLX .. 1.010 |=================================
n2-highcpu-8 CLX ... 1.000 |=================================
t2d-standard-8 AMD . 1.553 |===================================================
c2d-highcpu-8 AMD .. 1.475 |================================================
Geometric Mean Of Python Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.349 |===================================================
c2-standard-8 CLX .. 1.174 |============================================
n2-standard-8 CLX .. 1.077 |=========================================
n2-highcpu-8 CLX ... 1.070 |========================================
t2d-standard-8 AMD . 1.267 |================================================
c2d-highcpu-8 AMD .. 1.000 |======================================
Geometric Mean Of Renderers Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 2.000 |===================================================
c2-standard-8 CLX .. 1.500 |======================================
n2-standard-8 CLX .. 1.000 |==========================
n2-highcpu-8 CLX ... 1.000 |==========================
t2d-standard-8 AMD . 2.000 |===================================================
c2d-highcpu-8 AMD .. 1.500 |======================================
Geometric Mean Of Scientific Computing Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.421 |======================================
c2-standard-8 CLX .. 1.089 |=============================
n2-standard-8 CLX .. 1.000 |===========================
n2-highcpu-8 CLX ... 1.015 |===========================
t2d-standard-8 AMD . 1.894 |===================================================
c2d-highcpu-8 AMD .. 1.369 |=====================================
Geometric Mean Of Server Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.825 |==================================================
c2-standard-8 CLX .. 1.127 |===============================
n2-standard-8 CLX .. 1.018 |============================
n2-highcpu-8 CLX ... 1.000 |===========================
t2d-standard-8 AMD . 1.875 |===================================================
c2d-highcpu-8 AMD .. 1.365 |=====================================
Geometric Mean Of Server CPU Tests
Result Composite - Google Cloud c3 Sapphire Rapids vs. AMD Milan
Relative Performance > Higher Is Better
c3-highcpu-8 SPR ... 1.536 |==============================================
c2-standard-8 CLX .. 1.116 |=================================
n2-standard-8 CLX .. 1.001 |==============================
n2-highcpu-8 CLX ... 1.000 |==============================
t2d-standard-8 AMD . 1.699 |===================================================
c2d-highcpu-8 AMD .. 1.286 |=======================================
7-Zip Compression 22.01
Performance / Cost - Test: Compression Rating
MIPS Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 8826500.00 |================
c2-standard-8 CLX .. 7747250.00 |==============
n2-standard-8 CLX .. 7015500.00 |=============
n2-highcpu-8 CLX ... 9364333.33 |=================
t2d-standard-8 AMD . 24224500.00 |=============================================
c2d-highcpu-8 AMD .. 19601500.00 |====================================
Blender 3.4
Performance / Cost - Blend File: BMW27 - Compute: CPU-Only
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR . 9.457 |=====================================================
BRL-CAD 7.34
Performance / Cost - VGR Performance Metric
VGR Performance Metric Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1076848.49 |==========================================
c2-standard-8 CLX .. 541010.75 |=====================
n2-standard-8 CLX .. 512218.39 |====================
n2-highcpu-8 CLX ... 655235.29 |==========================
t2d-standard-8 AMD . 1166682.35 |==============================================
c2d-highcpu-8 AMD .. 923794.52 |====================================
CockroachDB 22.2
Performance / Cost - Workload: KV, 50% Reads - Concurrency: 128
ops/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1932160.00 |==============================================
c2-standard-8 CLX .. 1140375.00 |===========================
n2-standard-8 CLX .. 1058258.33 |=========================
n2-highcpu-8 CLX ... 1403900.00 |=================================
t2d-standard-8 AMD . 1911040.00 |=============================================
c2d-highcpu-8 AMD .. 1907600.00 |=============================================
CockroachDB 22.2
Performance / Cost - Workload: KV, 95% Reads - Concurrency: 128
ops/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 2496010.00 |==============================================
c2-standard-8 CLX .. 1348683.33 |=========================
n2-standard-8 CLX .. 1248066.67 |=======================
n2-highcpu-8 CLX ... 1653855.56 |==============================
t2d-standard-8 AMD . 2214880.00 |=========================================
c2d-highcpu-8 AMD .. 2285755.56 |==========================================
Embree 4.0.1
Performance / Cost - Binary: Pathtracer ISPC - Model: Crown
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 531.59 |==================================================
c2-standard-8 CLX .. 207.05 |===================
n2-standard-8 CLX .. 172.92 |================
n2-highcpu-8 CLX ... 231.43 |======================
t2d-standard-8 AMD . 499.35 |===============================================
c2d-highcpu-8 AMD .. 447.56 |==========================================
Embree 4.0.1
Performance / Cost - Binary: Pathtracer ISPC - Model: Asian Dragon
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 920.53 |==================================================
c2-standard-8 CLX .. 343.61 |===================
n2-standard-8 CLX .. 305.12 |=================
n2-highcpu-8 CLX ... 418.13 |=======================
t2d-standard-8 AMD . 713.03 |=======================================
c2d-highcpu-8 AMD .. 648.33 |===================================
Google Draco 1.5.6
Performance / Cost - Model: Lion
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 6.250 |======================================
c2-standard-8 CLX .. 7.437 |=============================================
n2-standard-8 CLX .. 8.355 |===================================================
n2-highcpu-8 CLX ... 8.391 |===================================================
t2d-standard-8 AMD . 6.073 |=====================================
c2d-highcpu-8 AMD .. 5.617 |==================================
Google Draco 1.5.6
Performance / Cost - Model: Church Facade
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 7.573 |===============
c2-standard-8 CLX .. 22.924 |==============================================
n2-standard-8 CLX .. 24.664 |==================================================
n2-highcpu-8 CLX ... 12.882 |==========================
t2d-standard-8 AMD . 8.754 |==================
c2d-highcpu-8 AMD .. 7.442 |===============
GROMACS 2023
Performance / Cost - Implementation: MPI CPU - Input: water_GMX50_bare
Ns Per Day Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 35.32 |=========================
c2-standard-8 CLX .. 16.54 |============
n2-standard-8 CLX .. 14.67 |==========
n2-highcpu-8 CLX ... 19.59 |==============
t2d-standard-8 AMD . 73.20 |===================================================
c2d-highcpu-8 AMD .. 32.86 |=======================
Intel Open Image Denoise 1.4.0
Performance / Cost - Run: RT.hdr_alb_nrm.3840x2160
Images / Sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 20.00 |===============================
c2-standard-8 CLX .. 14.67 |======================
n2-standard-8 CLX .. 12.50 |===================
n2-highcpu-8 CLX ... 16.67 |==========================
t2d-standard-8 AMD . 33.33 |===================================================
c2d-highcpu-8 AMD .. 10.67 |================
Intel Open Image Denoise 1.4.0
Performance / Cost - Run: RTLightmap.hdr.4096x4096
Images / Sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 4.800 |===============================
c2-standard-8 CLX .. 3.438 |======================
n2-standard-8 CLX .. 2.941 |===================
n2-highcpu-8 CLX ... 3.600 |=======================
t2d-standard-8 AMD . 7.895 |===================================================
c2d-highcpu-8 AMD .. 2.667 |=================
John The Ripper 2023.03.14
Performance / Cost - Test: bcrypt
Real C/S Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 2310666.67 |============================
c2-standard-8 CLX .. 2229000.00 |===========================
n2-standard-8 CLX .. 1990000.00 |========================
n2-highcpu-8 CLX ... 2981000.00 |====================================
t2d-standard-8 AMD . 3528333.33 |===========================================
c2d-highcpu-8 AMD .. 3810000.00 |==============================================
John The Ripper 2023.03.14
Performance / Cost - Test: Blowfish
Real C/S Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 2310000.00 |============================
c2-standard-8 CLX .. 2228000.00 |===========================
n2-standard-8 CLX .. 1989333.33 |========================
n2-highcpu-8 CLX ... 2979500.00 |====================================
t2d-standard-8 AMD . 3537000.00 |===========================================
c2d-highcpu-8 AMD .. 3809500.00 |==============================================
John The Ripper 2023.03.14
Performance / Cost - Test: MD5
Real C/S Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 127618833.33 |============================================
c2-standard-8 CLX .. 97649428.57 |==================================
n2-standard-8 CLX .. 99939500.00 |==================================
n2-highcpu-8 CLX ... 119794200.00 |=========================================
t2d-standard-8 AMD . 123035000.00 |==========================================
c2d-highcpu-8 AMD .. 87317000.00 |==============================
LeelaChessZero 0.28
Performance / Cost - Backend: BLAS
Nodes Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 37411.77 |================================================
c2-standard-8 CLX .. 21642.86 |============================
n2-standard-8 CLX .. 20769.23 |===========================
n2-highcpu-8 CLX ... 26551.72 |==================================
t2d-standard-8 AMD . 35352.94 |=============================================
c2d-highcpu-8 AMD .. 28533.33 |=====================================
LeelaChessZero 0.28
Performance / Cost - Backend: Eigen
Nodes Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 35911.77 |================================================
c2-standard-8 CLX .. 21476.19 |=============================
n2-standard-8 CLX .. 20717.95 |============================
n2-highcpu-8 CLX ... 27620.69 |=====================================
t2d-standard-8 AMD . 30558.82 |=========================================
c2d-highcpu-8 AMD .. 24766.67 |=================================
MariaDB 11.0.1
Performance / Cost - Clients: 2048
Queries Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 11066.67 |================================================
c2-standard-8 CLX .. 5166.67 |======================
n2-standard-8 CLX .. 4313.73 |===================
n2-highcpu-8 CLX ... 5945.95 |==========================
t2d-standard-8 AMD . 9406.25 |=========================================
c2d-highcpu-8 AMD .. 10892.86 |===============================================
MariaDB 11.0.1
Performance / Cost - Clients: 4096
Queries Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 9906.25 |=================================================
c2-standard-8 CLX .. 4647.06 |=======================
n2-standard-8 CLX .. 3833.33 |===================
n2-highcpu-8 CLX ... 5487.18 |===========================
t2d-standard-8 AMD . 8000.00 |========================================
c2d-highcpu-8 AMD .. 9733.33 |================================================
Memcached 1.6.18
Performance / Cost - Set To Get Ratio: 1:10
Ops/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 174157855.00 |======================================
c2-standard-8 CLX .. 89465441.25 |====================
n2-standard-8 CLX .. 91673371.43 |====================
n2-highcpu-8 CLX ... 122948524.00 |===========================
t2d-standard-8 AMD . 200087903.33 |============================================
c2d-highcpu-8 AMD .. 168566154.00 |=====================================
Memcached 1.6.18
Performance / Cost - Set To Get Ratio: 1:100
Ops/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 171822880.00 |=====================================
c2-standard-8 CLX .. 87786491.25 |===================
n2-standard-8 CLX .. 89282951.43 |===================
n2-highcpu-8 CLX ... 120872674.00 |==========================
t2d-standard-8 AMD . 204091780.00 |============================================
c2d-highcpu-8 AMD .. 167081012.00 |====================================
miniBUDE 20210901
Performance / Cost - Implementation: OpenMP - Input Deck: BM1
GFInst/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 18865.10 |==================
c2-standard-8 CLX .. 10186.53 |==========
n2-standard-8 CLX .. 8371.94 |========
n2-highcpu-8 CLX ... 11162.50 |===========
t2d-standard-8 AMD . 50675.50 |================================================
c2d-highcpu-8 AMD .. 20675.56 |====================
miniBUDE 20210901
Performance / Cost - Implementation: OpenMP - Input Deck: BM1
Billion Interactions/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 754.60 |==================
c2-standard-8 CLX .. 407.47 |==========
n2-standard-8 CLX .. 334.88 |========
n2-highcpu-8 CLX ... 446.50 |===========
t2d-standard-8 AMD . 2027.00 |=================================================
c2d-highcpu-8 AMD .. 827.00 |====================
NAMD 2.14
Performance / Cost - ATPase Simulation - 327,506 Atoms
days/ns x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.054 |====================
c2-standard-8 CLX .. 0.116 |===========================================
n2-standard-8 CLX .. 0.137 |===================================================
n2-highcpu-8 CLX ... 0.101 |======================================
t2d-standard-8 AMD . 0.044 |================
c2d-highcpu-8 AMD .. 0.080 |==============================
nekRS 22.0
Performance / Cost - Input: TurboPipe Periodic
FLOP/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 393178205128.21 |=======================
c2-standard-8 CLX .. 216385919543.10 |=============
n2-standard-8 CLX .. 194606497177.97 |===========
t2d-standard-8 AMD . 702695977017.24 |=========================================
c2d-highcpu-8 AMD .. 352145454545.46 |=====================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 757.46 |==============================
c2-standard-8 CLX .. 418.69 |=================
n2-standard-8 CLX .. 443.12 |==================
n2-highcpu-8 CLX ... 525.36 |=====================
t2d-standard-8 AMD . 1232.22 |=================================================
c2d-highcpu-8 AMD .. 854.85 |==================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 2.640 |============================
c2-standard-8 CLX .. 4.777 |===================================================
n2-standard-8 CLX .. 4.513 |================================================
n2-highcpu-8 CLX ... 3.807 |=========================================
t2d-standard-8 AMD . 3.227 |==================================
c2d-highcpu-8 AMD .. 2.338 |=========================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 4791.93 |=================================================
c2-standard-8 CLX .. 2652.98 |===========================
n2-standard-8 CLX .. 3072.42 |===============================
n2-highcpu-8 CLX ... 3778.38 |=======================================
t2d-standard-8 AMD . 4761.66 |=================================================
c2d-highcpu-8 AMD .. 3386.65 |===================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.417 |=========================
c2-standard-8 CLX .. 0.753 |==============================================
n2-standard-8 CLX .. 0.650 |========================================
n2-highcpu-8 CLX ... 0.529 |================================
t2d-standard-8 AMD . 0.838 |===================================================
c2d-highcpu-8 AMD .. 0.590 |====================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 16218.68 |===========================================
c2-standard-8 CLX .. 14997.85 |========================================
n2-standard-8 CLX .. 13739.75 |====================================
n2-highcpu-8 CLX ... 18171.97 |================================================
t2d-standard-8 AMD . 17213.78 |=============================================
c2d-highcpu-8 AMD .. 12708.07 |==================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.123 |===========================
c2-standard-8 CLX .. 0.133 |=============================
n2-standard-8 CLX .. 0.145 |================================
n2-highcpu-8 CLX ... 0.110 |========================
t2d-standard-8 AMD . 0.232 |===================================================
c2d-highcpu-8 AMD .. 0.157 |===================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 8276.60 |===============================
c2-standard-8 CLX .. 5835.66 |======================
n2-standard-8 CLX .. 5194.72 |===================
n2-highcpu-8 CLX ... 8564.57 |================================
t2d-standard-8 AMD . 12838.53 |================================================
c2d-highcpu-8 AMD .. 9496.50 |====================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.242 |================================
c2-standard-8 CLX .. 0.343 |=============================================
n2-standard-8 CLX .. 0.385 |===================================================
n2-highcpu-8 CLX ... 0.233 |===============================
t2d-standard-8 AMD . 0.311 |=========================================
c2d-highcpu-8 AMD .. 0.210 |============================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1307.44 |=======================================
c2-standard-8 CLX .. 884.19 |===========================
n2-standard-8 CLX .. 821.30 |=========================
n2-highcpu-8 CLX ... 1129.62 |==================================
t2d-standard-8 AMD . 1633.08 |=================================================
c2d-highcpu-8 AMD .. 1146.05 |==================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 1.529 |================================
c2-standard-8 CLX .. 2.262 |===============================================
n2-standard-8 CLX .. 2.435 |===================================================
n2-highcpu-8 CLX ... 1.770 |=====================================
t2d-standard-8 AMD . 2.445 |===================================================
c2d-highcpu-8 AMD .. 1.745 |====================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3243.88 |==============================
c2-standard-8 CLX .. 2332.20 |======================
n2-standard-8 CLX .. 2124.05 |====================
n2-highcpu-8 CLX ... 2576.18 |========================
t2d-standard-8 AMD . 5211.84 |=================================================
c2d-highcpu-8 AMD .. 3588.73 |==================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.616 |=================================
c2-standard-8 CLX .. 0.857 |==============================================
n2-standard-8 CLX .. 0.941 |===================================================
n2-highcpu-8 CLX ... 0.776 |==========================================
t2d-standard-8 AMD . 0.766 |==========================================
c2d-highcpu-8 AMD .. 0.557 |==============================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
items/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 753.86 |==============================
c2-standard-8 CLX .. 489.20 |===================
n2-standard-8 CLX .. 443.28 |==================
n2-highcpu-8 CLX ... 531.40 |=====================
t2d-standard-8 AMD . 1239.12 |=================================================
c2d-highcpu-8 AMD .. 847.73 |==================================
Neural Magic DeepSparse 1.3.2
Performance / Cost - Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream
ms/batch x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 2.653 |==============================
c2-standard-8 CLX .. 4.088 |==============================================
n2-standard-8 CLX .. 4.512 |===================================================
n2-highcpu-8 CLX ... 3.763 |===========================================
t2d-standard-8 AMD . 3.215 |====================================
c2d-highcpu-8 AMD .. 2.354 |===========================
nginx 1.23.2
Performance / Cost - Connections: 100
Requests Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 4034483.33 |==========================================
c2-standard-8 CLX .. 2514828.00 |==========================
n2-standard-8 CLX .. 2283019.00 |========================
n2-highcpu-8 CLX ... 3206342.86 |=================================
t2d-standard-8 AMD . 4407221.25 |==============================================
c2d-highcpu-8 AMD .. 3574265.71 |=====================================
nginx 1.23.2
Performance / Cost - Connections: 200
Requests Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3955788.89 |=========================================
c2-standard-8 CLX .. 2469591.00 |==========================
n2-standard-8 CLX .. 2257018.00 |========================
n2-highcpu-8 CLX ... 3061631.43 |================================
t2d-standard-8 AMD . 4403130.00 |==============================================
c2d-highcpu-8 AMD .. 3525290.00 |=====================================
nginx 1.23.2
Performance / Cost - Connections: 500
Requests Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3852516.67 |========================================
c2-standard-8 CLX .. 2195717.00 |=======================
n2-standard-8 CLX .. 2014610.00 |=====================
n2-highcpu-8 CLX ... 2829514.29 |==============================
t2d-standard-8 AMD . 4390907.50 |==============================================
c2d-highcpu-8 AMD .. 3466472.86 |====================================
nginx 1.23.2
Performance / Cost - Connections: 1000
Requests Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3568731.11 |=======================================
c2-standard-8 CLX .. 2144627.00 |=======================
n2-standard-8 CLX .. 1968360.00 |=====================
n2-highcpu-8 CLX ... 2774011.43 |==============================
t2d-standard-8 AMD . 4259952.50 |==============================================
c2d-highcpu-8 AMD .. 3394067.14 |=====================================
nginx 1.23.2
Performance / Cost - Connections: 4000
Requests Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3646083.33 |========================================
c2-standard-8 CLX .. 2159494.00 |========================
n2-standard-8 CLX .. 1994756.00 |======================
n2-highcpu-8 CLX ... 2782472.86 |==============================
t2d-standard-8 AMD . 4199096.25 |==============================================
c2d-highcpu-8 AMD .. 3392502.86 |=====================================
oneDNN 3.0
Performance / Cost - Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR .. 0.002 |==
c2-standard-8 CLX . 0.039 |==============================================
n2-standard-8 CLX . 0.044 |====================================================
n2-highcpu-8 CLX .. 0.022 |==========================
oneDNN 3.0
Performance / Cost - Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR .. 0.005 |=============================
c2-standard-8 CLX . 0.008 |==============================================
n2-standard-8 CLX . 0.009 |====================================================
n2-highcpu-8 CLX .. 0.009 |====================================================
oneDNN 3.0
Performance / Cost - Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR .. 0.004 |=====
c2-standard-8 CLX . 0.034 |=============================================
n2-standard-8 CLX . 0.039 |====================================================
oneDNN 3.0
Performance / Cost - Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR .. 0.003 |=
c2-standard-8 CLX . 0.106 |==============================================
n2-standard-8 CLX . 0.121 |====================================================
n2-highcpu-8 CLX .. 0.121 |====================================================
oneDNN 3.0
Performance / Cost - Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 41.95 |================================
c2-standard-8 CLX .. 63.45 |================================================
n2-standard-8 CLX .. 66.87 |===================================================
n2-highcpu-8 CLX ... 48.90 |=====================================
t2d-standard-8 AMD . 36.04 |===========================
c2d-highcpu-8 AMD .. 52.03 |========================================
oneDNN 3.0
Performance / Cost - Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 18.70 |================================
c2-standard-8 CLX .. 29.98 |===================================================
n2-standard-8 CLX .. 28.55 |=================================================
n2-highcpu-8 CLX ... 22.10 |======================================
t2d-standard-8 AMD . 16.86 |=============================
c2d-highcpu-8 AMD .. 23.35 |========================================
oneDNN 3.0
Performance / Cost - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR .. 0.001 |======
c2-standard-8 CLX . 0.007 |========================================
n2-standard-8 CLX . 0.009 |====================================================
n2-highcpu-8 CLX .. 0.008 |==============================================
OpenCV 4.7
Performance / Cost - Test: Core
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 698.98 |=======
c2-standard-8 CLX .. 2427.09 |=========================
n2-standard-8 CLX .. 4813.37 |=================================================
n2-highcpu-8 CLX ... 3539.38 |====================================
t2d-standard-8 AMD . 813.99 |========
c2d-highcpu-8 AMD .. 565.40 |======
OpenCV 4.7
Performance / Cost - Test: Graph API
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 4618.55 |=============================
c2-standard-8 CLX .. 6377.02 |========================================
n2-standard-8 CLX .. 7750.25 |=================================================
n2-highcpu-8 CLX ... 6650.73 |==========================================
t2d-standard-8 AMD . 4204.28 |===========================
c2d-highcpu-8 AMD .. 3390.99 |=====================
OpenCV 4.7
Performance / Cost - Test: Stitching
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 4509.96 |==========================
c2-standard-8 CLX .. 7274.16 |===========================================
n2-standard-8 CLX .. 8354.79 |=================================================
n2-highcpu-8 CLX ... 6189.44 |====================================
t2d-standard-8 AMD . 3524.81 |=====================
c2d-highcpu-8 AMD .. 2998.40 |==================
OpenCV 4.7
Performance / Cost - Test: Image Processing
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 1537.96 |======================
c2-standard-8 CLX .. 2502.98 |====================================
n2-standard-8 CLX .. 3395.38 |=================================================
n2-highcpu-8 CLX ... 2419.79 |===================================
t2d-standard-8 AMD . 1258.51 |==================
c2d-highcpu-8 AMD .. 1176.18 |=================
OpenCV 4.7
Performance / Cost - Test: Object Detection
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 156.00 |=================
c2-standard-8 CLX .. 406.39 |=============================================
n2-standard-8 CLX .. 449.99 |==================================================
n2-highcpu-8 CLX ... 331.35 |=====================================
t2d-standard-8 AMD . 96.04 |===========
c2d-highcpu-8 AMD .. 113.00 |=============
OpenFOAM 10
Performance / Cost - Input: drivaerFastback, Small Mesh Size - Mesh Time
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 2.916 |=====================
c2-standard-8 CLX .. 6.414 |===============================================
n2-standard-8 CLX .. 6.940 |===================================================
n2-highcpu-8 CLX ... 4.850 |====================================
t2d-standard-8 AMD . 1.609 |============
c2d-highcpu-8 AMD .. 2.493 |==================
OpenFOAM 10
Performance / Cost - Input: drivaerFastback, Small Mesh Size - Execution Time
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 19.866 |========================
c2-standard-8 CLX .. 42.046 |==================================================
n2-standard-8 CLX .. 41.919 |==================================================
n2-highcpu-8 CLX ... 29.885 |====================================
t2d-standard-8 AMD . 9.713 |============
c2d-highcpu-8 AMD .. 15.750 |===================
OpenRadioss 2022.10.13
Performance / Cost - Model: Bumper Beam
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 8.798 |=====================
c2-standard-8 CLX .. 17.577 |===========================================
n2-standard-8 CLX .. 20.505 |==================================================
n2-highcpu-8 CLX ... 15.185 |=====================================
t2d-standard-8 AMD . 3.006 |=======
c2d-highcpu-8 AMD .. 6.864 |=================
OpenRadioss 2022.10.13
Performance / Cost - Model: Cell Phone Drop Test
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 4.834 |=====================
c2-standard-8 CLX .. 10.202 |============================================
n2-standard-8 CLX .. 11.600 |==================================================
n2-highcpu-8 CLX ... 8.329 |====================================
t2d-standard-8 AMD . 1.768 |========
c2d-highcpu-8 AMD .. 3.850 |=================
OpenRadioss 2022.10.13
Performance / Cost - Model: Bird Strike on Windshield
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 33.92 |========================
c2-standard-8 CLX .. 60.88 |===========================================
n2-standard-8 CLX .. 71.44 |===================================================
n2-highcpu-8 CLX ... 51.20 |=====================================
t2d-standard-8 AMD . 13.91 |==========
c2d-highcpu-8 AMD .. 32.34 |=======================
OpenRadioss 2022.10.13
Performance / Cost - Model: Rubber O-Ring Seal Installation
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 12.845 |=================
c2-standard-8 CLX .. 31.930 |==========================================
n2-standard-8 CLX .. 38.461 |==================================================
n2-highcpu-8 CLX ... 27.452 |====================================
t2d-standard-8 AMD . 4.176 |=====
c2d-highcpu-8 AMD .. 11.294 |===============
OpenSSL 3.1
Performance / Cost - Algorithm: SHA256
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 251992587470.59 |==========================
c2-standard-8 CLX .. 62771120476.19 |======
n2-standard-8 CLX .. 62027045947.37 |======
n2-highcpu-8 CLX ... 84095258071.43 |=========
t2d-standard-8 AMD . 404478101764.71 |=========================================
c2d-highcpu-8 AMD .. 388278658866.67 |=======================================
OpenSSL 3.1
Performance / Cost - Algorithm: SHA512
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 92268995294.12 |====================
c2-standard-8 CLX .. 69800725095.24 |===============
n2-standard-8 CLX .. 68850211947.37 |===============
n2-highcpu-8 CLX ... 93301694285.71 |=====================
t2d-standard-8 AMD . 185803224529.41 |=========================================
c2d-highcpu-8 AMD .. 124223471133.33 |===========================
OpenSSL 3.1
Performance / Cost - Algorithm: RSA4096
sign/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 343783.33 |===============================================
c2-standard-8 CLX .. 165228.57 |=======================
n2-standard-8 CLX .. 171433.33 |=======================
n2-highcpu-8 CLX ... 205760.00 |============================
t2d-standard-8 AMD . 294166.67 |========================================
c2d-highcpu-8 AMD .. 205140.00 |============================
OpenSSL 3.1
Performance / Cost - Algorithm: ChaCha20
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1299503390411.76 |==================================
c2-standard-8 CLX .. 1016514848238.10 |==========================
n2-standard-8 CLX .. 1002901590368.42 |==========================
n2-highcpu-8 CLX ... 1361820421642.86 |===================================
t2d-standard-8 AMD . 1548546288823.53 |========================================
c2d-highcpu-8 AMD .. 1151101186000.00 |==============================
OpenSSL 3.1
Performance / Cost - Algorithm: AES-128-GCM
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 3387886930764.71 |========================================
c2-standard-8 CLX .. 1106538695380.95 |=============
n2-standard-8 CLX .. 1092908234368.42 |=============
n2-highcpu-8 CLX ... 1482396441214.29 |==================
t2d-standard-8 AMD . 1996959727823.53 |========================
c2d-highcpu-8 AMD .. 1234828201533.33 |===============
OpenSSL 3.1
Performance / Cost - Algorithm: AES-256-GCM
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 2824021269000.00 |========================================
c2-standard-8 CLX .. 806904791904.76 |===========
n2-standard-8 CLX .. 796275684368.42 |===========
n2-highcpu-8 CLX ... 1080989637357.14 |===============
t2d-standard-8 AMD . 1849851336470.59 |==========================
c2d-highcpu-8 AMD .. 1131262576866.67 |================
OpenSSL 3.1
Performance / Cost - Algorithm: ChaCha20-Poly1305
byte/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 939457714117.65 |====================================
c2-standard-8 CLX .. 519959611571.43 |====================
n2-standard-8 CLX .. 504027018578.95 |===================
n2-highcpu-8 CLX ... 683391021428.57 |==========================
t2d-standard-8 AMD . 1034858828058.82 |========================================
c2d-highcpu-8 AMD .. 783955097133.33 |==============================
OpenVKL 1.3.1
Performance / Cost - Benchmark: vklBenchmark ISPC
Items / Sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 2450.00 |====================================
c2-standard-8 CLX .. 1458.33 |=====================
n2-standard-8 CLX .. 1369.57 |====================
n2-highcpu-8 CLX ... 3375.00 |=================================================
t2d-standard-8 AMD . 1973.68 |=============================
c2d-highcpu-8 AMD .. 1823.53 |==========================
OSPRay Studio 0.11
Performance / Cost - Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 209.52 |===================
c2-standard-8 CLX .. 432.75 |========================================
n2-standard-8 CLX .. 538.26 |==================================================
n2-highcpu-8 CLX ... 387.12 |====================================
t2d-standard-8 AMD . 278.37 |==========================
c2d-highcpu-8 AMD .. 282.33 |==========================
OSPRay Studio 0.11
Performance / Cost - Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 309.83 |==========================
c2-standard-8 CLX .. 606.27 |==================================================
n2-standard-8 CLX .. 490.70 |========================================
n2-highcpu-8 CLX ... 366.90 |==============================
t2d-standard-8 AMD . 393.89 |================================
c2d-highcpu-8 AMD .. 372.27 |===============================
PostgreSQL 15
Performance / Cost - Scaling Factor: 100 - Clients: 800 - Mode: Read Only
TPS Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 22281571.43 |========================================
c2-standard-8 CLX .. 10195117.65 |==================
n2-standard-8 CLX .. 9891687.50 |==================
n2-highcpu-8 CLX ... 12674416.67 |=======================
t2d-standard-8 AMD . 24873307.69 |=============================================
c2d-highcpu-8 AMD .. 20790833.33 |======================================
PostgreSQL 15
Performance / Cost - Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.036 |=======================
c2-standard-8 CLX .. 0.078 |=================================================
n2-standard-8 CLX .. 0.081 |===================================================
n2-highcpu-8 CLX ... 0.063 |========================================
t2d-standard-8 AMD . 0.032 |====================
c2d-highcpu-8 AMD .. 0.038 |========================
PostgreSQL 15
Performance / Cost - Scaling Factor: 100 - Clients: 1000 - Mode: Read Only
TPS Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 20980357.14 |======================================
c2-standard-8 CLX .. 9979000.00 |==================
n2-standard-8 CLX .. 9699750.00 |==================
n2-highcpu-8 CLX ... 12149166.67 |======================
t2d-standard-8 AMD . 24679000.00 |=============================================
c2d-highcpu-8 AMD .. 20349833.33 |=====================================
PostgreSQL 15
Performance / Cost - Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency
ms x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.048 |========================
c2-standard-8 CLX .. 0.100 |==================================================
n2-standard-8 CLX .. 0.103 |===================================================
n2-highcpu-8 CLX ... 0.082 |=========================================
t2d-standard-8 AMD . 0.041 |====================
c2d-highcpu-8 AMD .. 0.049 |========================
SPECFEM3D 4.0
Performance / Cost - Model: Mount St. Helens
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 1.953 |==================================
c2-standard-8 CLX .. 2.471 |===========================================
n2-standard-8 CLX .. 2.924 |===================================================
n2-highcpu-8 CLX ... 2.084 |====================================
t2d-standard-8 AMD . 0.649 |===========
c2d-highcpu-8 AMD .. 1.382 |========================
SPECFEM3D 4.0
Performance / Cost - Model: Layered Halfspace
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 13.395 |====================================
c2-standard-8 CLX .. 16.489 |============================================
n2-standard-8 CLX .. 18.683 |==================================================
n2-highcpu-8 CLX ... 13.742 |=====================================
t2d-standard-8 AMD . 4.685 |=============
c2d-highcpu-8 AMD .. 8.771 |=======================
SPECFEM3D 4.0
Performance / Cost - Model: Tomographic Model
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 2.015 |==================================
c2-standard-8 CLX .. 2.716 |==============================================
n2-standard-8 CLX .. 2.995 |===================================================
n2-highcpu-8 CLX ... 2.343 |========================================
t2d-standard-8 AMD . 0.654 |===========
c2d-highcpu-8 AMD .. 1.410 |========================
SPECFEM3D 4.0
Performance / Cost - Model: Homogeneous Halfspace
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 3.232 |================================
c2-standard-8 CLX .. 4.388 |============================================
n2-standard-8 CLX .. 5.133 |===================================================
n2-highcpu-8 CLX ... 3.875 |=======================================
t2d-standard-8 AMD . 1.023 |==========
c2d-highcpu-8 AMD .. 2.296 |=======================
SPECFEM3D 4.0
Performance / Cost - Model: Water-layered Halfspace
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 10.292 |===============================
c2-standard-8 CLX .. 14.608 |=============================================
n2-standard-8 CLX .. 16.346 |==================================================
n2-highcpu-8 CLX ... 11.600 |===================================
t2d-standard-8 AMD . 3.753 |===========
c2d-highcpu-8 AMD .. 7.186 |======================
TensorFlow 2.10
Performance / Cost - Device: CPU - Batch Size: 16 - Model: ResNet-50
images/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1183.33 |=================================================
c2-standard-8 CLX .. 833.13 |==================================
n2-standard-8 CLX .. 761.25 |================================
n2-highcpu-8 CLX ... 1015.00 |==========================================
t2d-standard-8 AMD . 671.25 |============================
c2d-highcpu-8 AMD .. 509.41 |=====================
TensorFlow 2.10
Performance / Cost - Device: CPU - Batch Size: 32 - Model: ResNet-50
images/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 649.13 |==================================================
c2-standard-8 CLX .. 470.33 |====================================
n2-standard-8 CLX .. 431.67 |=================================
n2-highcpu-8 CLX ... 590.46 |=============================================
t2d-standard-8 AMD . 331.56 |==========================
c2d-highcpu-8 AMD .. 244.29 |===================
TensorFlow 2.10
Performance / Cost - Device: CPU - Batch Size: 64 - Model: ResNet-50
images/sec Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 364.88 |==================================================
c2-standard-8 CLX .. 264.11 |====================================
n2-standard-8 CLX .. 245.36 |==================================
n2-highcpu-8 CLX ... 326.19 |=============================================
t2d-standard-8 AMD . 163.75 |======================
c2d-highcpu-8 AMD .. 120.86 |=================
Timed FFmpeg Compilation 6.0
Performance / Cost - Time To Compile
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 1.325 |==========================
c2-standard-8 CLX .. 2.225 |===========================================
n2-standard-8 CLX .. 2.633 |===================================================
n2-highcpu-8 CLX ... 1.859 |====================================
t2d-standard-8 AMD . 0.645 |============
c2d-highcpu-8 AMD .. 1.196 |=======================
Timed Linux Kernel Compilation 6.1
Performance / Cost - Build: defconfig
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 5.630 |=========================
c2-standard-8 CLX .. 9.849 |============================================
n2-standard-8 CLX .. 11.283 |==================================================
n2-highcpu-8 CLX ... 8.522 |======================================
t2d-standard-8 AMD . 3.140 |==============
c2d-highcpu-8 AMD .. 5.705 |=========================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 4K - Video Preset: Very Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 873.75 |====================
c2-standard-8 CLX .. 474.17 |===========
n2-standard-8 CLX .. 382.31 |=========
n2-highcpu-8 CLX ... 493.00 |===========
t2d-standard-8 AMD . 2136.00 |=================================================
c2d-highcpu-8 AMD .. 1316.67 |==============================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 4K - Video Preset: Super Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 935.00 |====================
c2-standard-8 CLX .. 502.50 |===========
n2-standard-8 CLX .. 440.00 |=========
n2-highcpu-8 CLX ... 574.44 |============
t2d-standard-8 AMD . 2272.00 |=================================================
c2d-highcpu-8 AMD .. 1356.67 |=============================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1520.00 |======================
c2-standard-8 CLX .. 827.78 |============
n2-standard-8 CLX .. 651.00 |=========
n2-highcpu-8 CLX ... 915.71 |=============
t2d-standard-8 AMD . 3417.50 |=================================================
c2d-highcpu-8 AMD .. 1964.00 |============================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 1080p - Video Preset: Very Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 16195.00 |================
c2-standard-8 CLX .. 8766.67 |=========
n2-standard-8 CLX .. 7603.33 |=======
n2-highcpu-8 CLX ... 11435.00 |===========
t2d-standard-8 AMD . 49190.00 |================================================
c2d-highcpu-8 AMD .. 36450.00 |====================================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 1080p - Video Preset: Super Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 17250.00 |================
c2-standard-8 CLX .. 9240.00 |========
n2-standard-8 CLX .. 8050.00 |=======
n2-highcpu-8 CLX ... 12075.00 |===========
t2d-standard-8 AMD . 52250.00 |================================================
c2d-highcpu-8 AMD .. 37790.00 |===================================
uvg266 0.4.1
Performance / Cost - Video Input: Bosphorus 1080p - Video Preset: Ultra Fast
Frames Per Second Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 42240.00 |=================================
c2-standard-8 CLX .. 17235.00 |=============
n2-standard-8 CLX .. 15025.00 |============
n2-highcpu-8 CLX ... 14985.00 |============
t2d-standard-8 AMD . 61850.00 |================================================
c2d-highcpu-8 AMD .. 45530.00 |===================================
Xcompact3d Incompact3d 2021-03-11
Performance / Cost - Input: input.i3d 129 Cells Per Direction
Seconds x Dollar < Lower Is Better
c3-highcpu-8 SPR ... 0.097 |===============================
c2-standard-8 CLX .. 0.151 |================================================
n2-standard-8 CLX .. 0.162 |===================================================
n2-highcpu-8 CLX ... 0.120 |======================================
t2d-standard-8 AMD . 0.043 |==============
c2d-highcpu-8 AMD .. 0.053 |=================
Zstd Compression 1.5.4
Performance / Cost - Compression Level: 19 - Compression Speed
MB/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1471.43 |=======================================
c2-standard-8 CLX .. 1026.67 |===========================
n2-standard-8 CLX .. 915.56 |========================
n2-highcpu-8 CLX ... 1184.29 |================================
t2d-standard-8 AMD . 1557.14 |==========================================
c2d-highcpu-8 AMD .. 1833.33 |=================================================
Zstd Compression 1.5.4
Performance / Cost - Compression Level: 19 - Decompression Speed
MB/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 129314.29 |=============================
c2-standard-8 CLX .. 77900.00 |==================
n2-standard-8 CLX .. 72166.67 |================
n2-highcpu-8 CLX ... 91542.86 |=====================
t2d-standard-8 AMD . 161285.71 |====================================
c2d-highcpu-8 AMD .. 208933.33 |===============================================
Zstd Compression 1.5.4
Performance / Cost - Compression Level: 19, Long Mode - Compression Speed
MB/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 1083.33 |========================================
c2-standard-8 CLX .. 732.50 |===========================
n2-standard-8 CLX .. 641.25 |=======================
n2-highcpu-8 CLX ... 840.00 |===============================
t2d-standard-8 AMD . 774.29 |============================
c2d-highcpu-8 AMD .. 1342.00 |=================================================
Zstd Compression 1.5.4
Performance / Cost - Compression Level: 19, Long Mode - Decompression Speed
MB/s Per Dollar > Higher Is Better
c3-highcpu-8 SPR ... 151200.00 |==============================
c2-standard-8 CLX .. 89225.00 |==================
n2-standard-8 CLX .. 82537.50 |================
n2-highcpu-8 CLX ... 108333.33 |======================
t2d-standard-8 AMD . 153685.71 |===============================
c2d-highcpu-8 AMD .. 236760.00 |===============================================