GCE c3d-standard-60
amazon testing on Ubuntu 22.04 via the Phoronix Test Suite.


c3d-standard-60 AMD Genoa: 

	Processor: AMD EPYC 9B14 (30 Cores / 60 Threads), Motherboard: Google Compute Engine c3d-standard-60, Chipset: Intel 440FX 82441FX PMC, Memory: 240GB, Disk: 215GB nvme_card-pd, Network: Google Compute Engine Virtual

	OS: Ubuntu 22.04, Kernel: 6.2.0-1014-gcp (x86_64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: KVM

t2d-standard-60 AMD Milan: 

	Processor: AMD EPYC 7B13 (60 Cores), Motherboard: Google Compute Engine t2d-standard-60, Chipset: Intel 440FX 82441FX PMC, Memory: 240GB, Disk: 215GB PersistentDisk, Network: Red Hat Virtio device

	OS: Ubuntu 22.04, Kernel: 6.2.0-1014-gcp (x86_64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: KVM

c6g.16xlarge: 

	Processor: ARMv8 Neoverse-N1 (64 Cores), Motherboard: Amazon EC2 c6g.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 128GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic

	OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: amazon

m7a.16xlarge: 

	Processor: AMD EPYC 9R14 (64 Cores), Motherboard: Amazon EC2 m7a.16xlarge (1.0 BIOS), Chipset: Intel 440FX 82441FX PMC, Memory: 256GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic

	OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (x86_64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: amazon


NAS Parallel Benchmarks 3.4
Test / Class: BT.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 96257.48  |====================
t2d-standard-60 AMD Milan . 122720.61 |=========================
c6g.16xlarge .............. 24229.14  |=====
m7a.16xlarge .............. 193219.12 |========================================


NAS Parallel Benchmarks 3.4
Test / Class: CG.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 19597.86 |===================
t2d-standard-60 AMD Milan . 16649.37 |================
c6g.16xlarge .............. 13343.35 |=============
m7a.16xlarge .............. 42007.57 |=========================================


NAS Parallel Benchmarks 3.4
Test / Class: EP.D
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 3783.60 |=====================
t2d-standard-60 AMD Milan . 4935.68 |============================
c6g.16xlarge .............. 2213.76 |============
m7a.16xlarge .............. 7501.76 |==========================================


NAS Parallel Benchmarks 3.4
Test / Class: FT.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 39647.47  |===============
t2d-standard-60 AMD Milan . 54846.18  |=====================
c6g.16xlarge .............. 21386.37  |========
m7a.16xlarge .............. 103413.02 |========================================


NAS Parallel Benchmarks 3.4
Test / Class: IS.D
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 2422.40 |=========================
t2d-standard-60 AMD Milan . 1752.62 |==================
c6g.16xlarge .............. 915.80  |=========
m7a.16xlarge .............. 4085.20 |==========================================


NAS Parallel Benchmarks 3.4
Test / Class: LU.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 73563.13  |==============
t2d-standard-60 AMD Milan . 94247.77  |==================
c6g.16xlarge .............. 18807.75  |====
m7a.16xlarge .............. 210544.87 |========================================


NAS Parallel Benchmarks 3.4
Test / Class: MG.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 42701.83  |==============
t2d-standard-60 AMD Milan . 47291.96  |================
c6g.16xlarge .............. 25661.04  |========
m7a.16xlarge .............. 121293.80 |========================================


NAS Parallel Benchmarks 3.4
Test / Class: SP.C
Total Mop/s > Higher Is Better
c3d-standard-60 AMD Genoa . 39919.71  |================
t2d-standard-60 AMD Milan . 43228.11  |=================
c6g.16xlarge .............. 9716.99   |====
m7a.16xlarge .............. 102392.40 |========================================


Rodinia 3.1
Test: OpenMP LavaMD
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 64.86 |============================================
t2d-standard-60 AMD Milan . 50.97 |===================================
c6g.16xlarge .............. 62.30 |==========================================
m7a.16xlarge .............. 43.29 |=============================


Rodinia 3.1
Test: OpenMP HotSpot3D
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 84.17 |==========================================
t2d-standard-60 AMD Milan . 88.54 |============================================
m7a.16xlarge .............. 74.78 |=====================================


Rodinia 3.1
Test: OpenMP Leukocyte
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 45.50 |============================================
t2d-standard-60 AMD Milan . 42.01 |=========================================
m7a.16xlarge .............. 34.66 |==================================


Rodinia 3.1
Test: OpenMP CFD Solver
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 10.025 |===========================================
t2d-standard-60 AMD Milan . 7.368  |================================
c6g.16xlarge .............. 5.983  |==========================
m7a.16xlarge .............. 6.480  |============================


Rodinia 3.1
Test: OpenMP Streamcluster
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 6.448  |====================
t2d-standard-60 AMD Milan . 6.423  |===================
c6g.16xlarge .............. 14.212 |===========================================
m7a.16xlarge .............. 5.930  |==================


Algebraic Multi-Grid Benchmark 1.2

Figure Of Merit > Higher Is Better
c3d-standard-60 AMD Genoa . 962889833  |====================
t2d-standard-60 AMD Milan . 920427767  |===================
c6g.16xlarge .............. 1032893667 |======================
m7a.16xlarge .............. 1843444333 |=======================================


libxsmm 2-1.17-3645
M N K: 32
GFLOPS/s > Higher Is Better
c3d-standard-60 AMD Genoa . 255.4 |=================
t2d-standard-60 AMD Milan . 289.2 |====================
c6g.16xlarge .............. 312.7 |=====================
m7a.16xlarge .............. 643.4 |============================================


libxsmm 2-1.17-3645
M N K: 64
GFLOPS/s > Higher Is Better
c3d-standard-60 AMD Genoa . 489.7  |==================
t2d-standard-60 AMD Milan . 554.2  |====================
c6g.16xlarge .............. 589.5  |=====================
m7a.16xlarge .............. 1201.8 |===========================================


Laghos 3.1
Test: Triple Point Problem
Major Kernels Total Rate > Higher Is Better
c3d-standard-60 AMD Genoa . 209.00 |========================================
t2d-standard-60 AMD Milan . 222.30 |===========================================
c6g.16xlarge .............. 179.52 |===================================
m7a.16xlarge .............. 218.86 |==========================================


Laghos 3.1
Test: Sedov Blast Wave, ube_922_hex.mesh
Major Kernels Total Rate > Higher Is Better
c3d-standard-60 AMD Genoa . 259.55 |===========================
t2d-standard-60 AMD Milan . 364.64 |======================================
c6g.16xlarge .............. 321.29 |==================================
m7a.16xlarge .............. 409.73 |===========================================


HeFFTe - Highly Efficient FFT for Exascale 2.3
Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128
GFLOP/s > Higher Is Better
c3d-standard-60 AMD Genoa . 88.63  |==============================
t2d-standard-60 AMD Milan . 109.68 |=====================================
c6g.16xlarge .............. 129.17 |===========================================
m7a.16xlarge .............. 121.51 |========================================


HeFFTe - Highly Efficient FFT for Exascale 2.3
Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128
GFLOP/s > Higher Is Better
c3d-standard-60 AMD Genoa . 148.58 |================================
t2d-standard-60 AMD Milan . 196.95 |==========================================
c6g.16xlarge .............. 202.45 |===========================================
m7a.16xlarge .............. 190.60 |========================================


HeFFTe - Highly Efficient FFT for Exascale 2.3
Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128
GFLOP/s > Higher Is Better
c3d-standard-60 AMD Genoa . 57.31 |===================================
t2d-standard-60 AMD Milan . 60.03 |=====================================
c6g.16xlarge .............. 32.36 |====================
m7a.16xlarge .............. 71.11 |============================================


HeFFTe - Highly Efficient FFT for Exascale 2.3
Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128
GFLOP/s > Higher Is Better
c3d-standard-60 AMD Genoa . 93.60  |================================
t2d-standard-60 AMD Milan . 106.03 |=====================================
c6g.16xlarge .............. 79.02  |===========================
m7a.16xlarge .............. 124.36 |===========================================


Xcompact3d Incompact3d 2021-03-11
Input: input.i3d 129 Cells Per Direction
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 5.87157885 |=======================================
t2d-standard-60 AMD Milan . 5.63057327 |=====================================
c6g.16xlarge .............. 5.61811686 |=====================================
m7a.16xlarge .............. 2.89602661 |===================


Xcompact3d Incompact3d 2021-03-11
Input: input.i3d 193 Cells Per Direction
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 28.02 |============================================
t2d-standard-60 AMD Milan . 24.57 |=======================================
c6g.16xlarge .............. 25.87 |=========================================
m7a.16xlarge .............. 11.59 |==================


OpenRadioss 2023.09.15
Model: Bumper Beam
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 92.87 |============================================
t2d-standard-60 AMD Milan . 75.68 |====================================
m7a.16xlarge .............. 66.27 |===============================


OpenRadioss 2023.09.15
Model: Chrysler Neon 1M
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 337.70 |===========================================
t2d-standard-60 AMD Milan . 327.88 |==========================================
m7a.16xlarge .............. 190.79 |========================


OpenRadioss 2023.09.15
Model: Cell Phone Drop Test
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 38.82 |============================================
t2d-standard-60 AMD Milan . 30.12 |==================================
m7a.16xlarge .............. 26.14 |==============================


OpenRadioss 2023.09.15
Model: Bird Strike on Windshield
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 147.31 |===========================================
t2d-standard-60 AMD Milan . 123.61 |====================================
m7a.16xlarge .............. 115.96 |==================================


OpenRadioss 2023.09.15
Model: Rubber O-Ring Seal Installation
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 89.65 |============================================
t2d-standard-60 AMD Milan . 72.06 |===================================
m7a.16xlarge .............. 59.18 |=============================


OpenRadioss 2023.09.15
Model: INIVOL and Fluid Structure Interaction Drop Container
Seconds < Lower Is Better


Remhos 1.0
Test: Sample Remap Example
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 33.36 |============================================
t2d-standard-60 AMD Milan . 16.33 |======================
c6g.16xlarge .............. 20.82 |===========================
m7a.16xlarge .............. 13.87 |==================


nekRS 23.0
Input: Kershaw
flops/rank > Higher Is Better
c3d-standard-60 AMD Genoa . 4289858333 |======================
t2d-standard-60 AMD Milan . 3681935833 |===================
c6g.16xlarge .............. 1758860000 |=========
m7a.16xlarge .............. 7667846667 |=======================================


nekRS 23.0
Input: TurboPipe Periodic
flops/rank > Higher Is Better
c3d-standard-60 AMD Genoa . 4723940000 |=======================================
t2d-standard-60 AMD Milan . 2730620000 |======================
c6g.16xlarge .............. 2221710000 |==================
m7a.16xlarge .............. 4774796667 |=======================================


LAMMPS Molecular Dynamics Simulator 23Jun2022
Model: 20k Atoms
ns/day > Higher Is Better
c3d-standard-60 AMD Genoa . 19.78 |============================
t2d-standard-60 AMD Milan . 26.73 |=====================================
c6g.16xlarge .............. 25.06 |===================================
m7a.16xlarge .............. 31.47 |============================================


LAMMPS Molecular Dynamics Simulator 23Jun2022
Model: Rhodopsin Protein
ns/day > Higher Is Better
c3d-standard-60 AMD Genoa . 17.42 |=======================
t2d-standard-60 AMD Milan . 27.83 |=====================================
c6g.16xlarge .............. 26.04 |===================================
m7a.16xlarge .............. 32.79 |============================================


Coremark 1.0
CoreMark Size 666 - Iterations Per Second
Iterations/Sec > Higher Is Better
c3d-standard-60 AMD Genoa . 1445843.52 |==========================
t2d-standard-60 AMD Milan . 1730658.45 |===============================
c6g.16xlarge .............. 1259870.72 |=======================
m7a.16xlarge .............. 2158639.27 |=======================================


7-Zip Compression 22.01
Test: Compression Rating
MIPS > Higher Is Better
c3d-standard-60 AMD Genoa . 271795 |===================================
t2d-standard-60 AMD Milan . 278973 |====================================
c6g.16xlarge .............. 239735 |===============================
m7a.16xlarge .............. 330633 |===========================================


7-Zip Compression 22.01
Test: Decompression Rating
MIPS > Higher Is Better
c3d-standard-60 AMD Genoa . 226211 |==================================
t2d-standard-60 AMD Milan . 247255 |======================================
c6g.16xlarge .............. 234046 |====================================
m7a.16xlarge .............. 282593 |===========================================


Stockfish 15
Total Time
Nodes Per Second > Higher Is Better
c3d-standard-60 AMD Genoa . 105894457 |===============================
t2d-standard-60 AMD Milan . 112958788 |=================================
c6g.16xlarge .............. 81807706  |========================
m7a.16xlarge .............. 135419169 |========================================


libavif avifenc 1.0
Encoder Speed: 0
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 78.07  |============
t2d-standard-60 AMD Milan . 78.35  |============
c6g.16xlarge .............. 270.07 |===========================================
m7a.16xlarge .............. 65.45  |==========


libavif avifenc 1.0
Encoder Speed: 2
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 41.54  |===========
t2d-standard-60 AMD Milan . 41.99  |===========
c6g.16xlarge .............. 167.95 |===========================================
m7a.16xlarge .............. 35.81  |=========


libavif avifenc 1.0
Encoder Speed: 6
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 3.250 |================================
t2d-standard-60 AMD Milan . 3.205 |================================
c6g.16xlarge .............. 4.467 |============================================
m7a.16xlarge .............. 2.649 |==========================


libavif avifenc 1.0
Encoder Speed: 6, Lossless
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 6.889 |==================================
t2d-standard-60 AMD Milan . 7.639 |======================================
c6g.16xlarge .............. 8.879 |============================================
m7a.16xlarge .............. 5.678 |============================


Timed Gem5 Compilation 21.2
Time To Compile
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 176.77 |==================================
t2d-standard-60 AMD Milan . 170.93 |=================================
c6g.16xlarge .............. 224.41 |===========================================
m7a.16xlarge .............. 153.80 |=============================


Timed Linux Kernel Compilation 6.1
Build: defconfig
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 33.40  |==============
c6g.16xlarge .............. 102.22 |===========================================
m7a.16xlarge .............. 27.71  |============


Timed Linux Kernel Compilation 6.1
Build: allmodconfig
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 333.35 |===================================
c6g.16xlarge .............. 409.10 |===========================================
m7a.16xlarge .............. 267.97 |============================


Timed Node.js Compilation 19.8.1
Time To Compile
Seconds < Lower Is Better
c3d-standard-60 AMD Genoa . 198.39 |==============================
t2d-standard-60 AMD Milan . 191.71 |=============================
c6g.16xlarge .............. 286.20 |===========================================
m7a.16xlarge .............. 154.44 |=======================


OpenSSL 3.1
Algorithm: SHA256
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 46211821313 |============================
t2d-standard-60 AMD Milan . 50884997103 |===============================
c6g.16xlarge .............. 42288513973 |==========================
m7a.16xlarge .............. 62253861197 |======================================


OpenSSL 3.1
Algorithm: SHA512
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 14702270573 |=====================
t2d-standard-60 AMD Milan . 22244804183 |================================
c6g.16xlarge .............. 14384917863 |=====================
m7a.16xlarge .............. 26481506820 |======================================


OpenSSL 3.1
Algorithm: RSA4096
sign/s > Higher Is Better
c3d-standard-60 AMD Genoa . 20079.5 |===========================
t2d-standard-60 AMD Milan . 12973.0 |=================
c6g.16xlarge .............. 2640.0  |====
m7a.16xlarge .............. 31583.8 |==========================================


OpenSSL 3.1
Algorithm: RSA4096
verify/s > Higher Is Better
c3d-standard-60 AMD Genoa . 493077.6 |====================
t2d-standard-60 AMD Milan . 860844.6 |===================================
c6g.16xlarge .............. 215683.2 |=========
m7a.16xlarge .............. 996017.5 |=========================================


OpenSSL 3.1
Algorithm: ChaCha20
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 173980949893 |=====================
t2d-standard-60 AMD Milan . 180249145770 |======================
c6g.16xlarge .............. 67324778360  |========
m7a.16xlarge .............. 308308045083 |=====================================


OpenSSL 3.1
Algorithm: AES-128-GCM
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 343095284440 |=====================
t2d-standard-60 AMD Milan . 234604082610 |===============
c6g.16xlarge .............. 158788510970 |==========
m7a.16xlarge .............. 592545362740 |=====================================


OpenSSL 3.1
Algorithm: AES-256-GCM
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 293328048497 |=====================
t2d-standard-60 AMD Milan . 216025967640 |===============
c6g.16xlarge .............. 129198197600 |=========
m7a.16xlarge .............. 522113080527 |=====================================


OpenSSL 3.1
Algorithm: ChaCha20-Poly1305
byte/s > Higher Is Better
c3d-standard-60 AMD Genoa . 123909304773 |=====================
t2d-standard-60 AMD Milan . 119647720337 |====================
c6g.16xlarge .............. 46715126487  |========
m7a.16xlarge .............. 216773475533 |=====================================


Apache IoTDB 1.2
Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400
point/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 33268158 |=================================
t2d-standard-60 AMD Milan . 33466804 |==================================
m7a.16xlarge .............. 40899210 |=========================================


Apache IoTDB 1.2
Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400
Average Latency < Lower Is Better
c3d-standard-60 AMD Genoa . 418.59 |===========================================
t2d-standard-60 AMD Milan . 415.08 |===========================================
m7a.16xlarge .............. 340.11 |===================================


Apache IoTDB 1.2
Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400
point/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 34762565 |================================
t2d-standard-60 AMD Milan . 34925899 |================================
m7a.16xlarge .............. 44502333 |=========================================


Apache IoTDB 1.2
Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400
Average Latency < Lower Is Better
c3d-standard-60 AMD Genoa . 623.89 |==========================================
t2d-standard-60 AMD Milan . 633.58 |===========================================
m7a.16xlarge .............. 521.98 |===================================


Apache IoTDB 1.2
Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400
point/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 34332237 |=================================
t2d-standard-60 AMD Milan . 34123810 |=================================
m7a.16xlarge .............. 42643903 |=========================================


Apache IoTDB 1.2
Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400
Average Latency < Lower Is Better
c3d-standard-60 AMD Genoa . 447.68 |===========================================
t2d-standard-60 AMD Milan . 433.96 |==========================================
m7a.16xlarge .............. 355.13 |==================================


Apache IoTDB 1.2
Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400
point/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 35359884 |================================
t2d-standard-60 AMD Milan . 35068557 |================================
m7a.16xlarge .............. 44699315 |=========================================


Apache IoTDB 1.2
Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400
Average Latency < Lower Is Better
c3d-standard-60 AMD Genoa . 682.12 |=========================================
t2d-standard-60 AMD Milan . 709.46 |===========================================
m7a.16xlarge .............. 598.49 |====================================


GROMACS 2023
Implementation: MPI CPU - Input: water_GMX50_bare
Ns Per Day > Higher Is Better
c3d-standard-60 AMD Genoa . 4.391 |=========================
t2d-standard-60 AMD Milan . 5.289 |==============================
c6g.16xlarge .............. 2.766 |================
m7a.16xlarge .............. 7.655 |============================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 800 - Mode: Read Only
TPS > Higher Is Better
t2d-standard-60 AMD Milan . 2003784 |=============================
c6g.16xlarge .............. 1043267 |===============
m7a.16xlarge .............. 2923009 |==========================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 1000 - Mode: Read Only
TPS > Higher Is Better
t2d-standard-60 AMD Milan . 2008186 |=============================
c6g.16xlarge .............. 975031  |==============
m7a.16xlarge .............. 2880940 |==========================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 800 - Mode: Read Write
TPS > Higher Is Better
t2d-standard-60 AMD Milan . 5682 |=============================================
c6g.16xlarge .............. 4784 |======================================
m7a.16xlarge .............. 5312 |==========================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 1000 - Mode: Read Write
TPS > Higher Is Better
t2d-standard-60 AMD Milan . 5793 |=============================================
c6g.16xlarge .............. 4776 |=====================================
m7a.16xlarge .............. 5300 |=========================================


TensorFlow 2.12
Device: CPU - Batch Size: 16 - Model: ResNet-50
images/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 50.99 |================================
t2d-standard-60 AMD Milan . 18.29 |============
m7a.16xlarge .............. 69.55 |============================================


TensorFlow 2.12
Device: CPU - Batch Size: 32 - Model: ResNet-50
images/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 62.74 |================================
t2d-standard-60 AMD Milan . 20.36 |==========
m7a.16xlarge .............. 87.19 |============================================


TensorFlow 2.12
Device: CPU - Batch Size: 64 - Model: ResNet-50
images/sec > Higher Is Better
c3d-standard-60 AMD Genoa . 69.68  |==============================
t2d-standard-60 AMD Milan . 20.90  |=========
m7a.16xlarge .............. 100.15 |===========================================


Blender 3.6
Blend File: BMW27 - Compute: CPU-Only
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 34.27 |============================================
m7a.16xlarge .............. 27.74 |====================================


Blender 3.6
Blend File: Classroom - Compute: CPU-Only
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 89.35 |============================================
m7a.16xlarge .............. 71.51 |===================================


Blender 3.6
Blend File: Fishy Cat - Compute: CPU-Only
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 45.22 |============================================
m7a.16xlarge .............. 37.12 |====================================


Blender 3.6
Blend File: Barbershop - Compute: CPU-Only
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 351.58 |===========================================
m7a.16xlarge .............. 276.23 |==================================


Blender 3.6
Blend File: Pabellon Barcelona - Compute: CPU-Only
Seconds < Lower Is Better
t2d-standard-60 AMD Milan . 112.64 |===========================================
m7a.16xlarge .............. 91.88  |===================================


OpenVINO 2023.1
Model: Face Detection FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 18.39 |==========================
t2d-standard-60 AMD Milan . 10.73 |===============
c6g.16xlarge .............. 0.10  |
m7a.16xlarge .............. 31.72 |============================================


OpenVINO 2023.1
Model: Face Detection FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 648.81  |===
t2d-standard-60 AMD Milan . 1393.56 |======
c6g.16xlarge .............. 9996.56 |==========================================
m7a.16xlarge .............. 503.38  |==


OpenVINO 2023.1
Model: Person Detection FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 142.75 |======================
t2d-standard-60 AMD Milan . 73.74  |===========
c6g.16xlarge .............. 1.06   |
m7a.16xlarge .............. 284.42 |===========================================


OpenVINO 2023.1
Model: Person Detection FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 83.99  |====
t2d-standard-60 AMD Milan . 208.47 |=========
c6g.16xlarge .............. 947.59 |===========================================
m7a.16xlarge .............. 56.21  |===


OpenVINO 2023.1
Model: Person Detection FP32 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 142.90 |======================
t2d-standard-60 AMD Milan . 78.96  |============
c6g.16xlarge .............. 1.06   |
m7a.16xlarge .............. 283.40 |===========================================


OpenVINO 2023.1
Model: Person Detection FP32 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 83.91  |====
t2d-standard-60 AMD Milan . 193.45 |=========
c6g.16xlarge .............. 947.86 |===========================================
m7a.16xlarge .............. 56.41  |===


OpenVINO 2023.1
Model: Vehicle Detection FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 1389.69 |========================
t2d-standard-60 AMD Milan . 368.39  |======
c6g.16xlarge .............. 6.53    |
m7a.16xlarge .............. 2417.34 |==========================================


OpenVINO 2023.1
Model: Vehicle Detection FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 8.62   |==
t2d-standard-60 AMD Milan . 40.67  |===========
c6g.16xlarge .............. 153.12 |===========================================
m7a.16xlarge .............. 6.60   |==


OpenVINO 2023.1
Model: Face Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 35.19 |=========================
t2d-standard-60 AMD Milan . 26.28 |===================
c6g.16xlarge .............. 0.04  |
m7a.16xlarge .............. 61.19 |============================================


OpenVINO 2023.1
Model: Face Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 340.29   |=
t2d-standard-60 AMD Milan . 568.77   |=
c6g.16xlarge .............. 22391.86 |=========================================
m7a.16xlarge .............. 261.11   |


OpenVINO 2023.1
Model: Face Detection Retail FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 4166.39 |========================
t2d-standard-60 AMD Milan . 1285.47 |=======
c6g.16xlarge .............. 20.79   |
m7a.16xlarge .............. 7222.10 |==========================================


OpenVINO 2023.1
Model: Face Detection Retail FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 2.87  |===
t2d-standard-60 AMD Milan . 11.65 |===========
c6g.16xlarge .............. 48.08 |============================================
m7a.16xlarge .............. 2.21  |==


OpenVINO 2023.1
Model: Road Segmentation ADAS FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 576.94  |=======================
t2d-standard-60 AMD Milan . 225.48  |=========
c6g.16xlarge .............. 2.61    |
m7a.16xlarge .............. 1049.66 |==========================================


OpenVINO 2023.1
Model: Road Segmentation ADAS FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 20.77  |==
t2d-standard-60 AMD Milan . 66.46  |=======
c6g.16xlarge .............. 382.47 |===========================================
m7a.16xlarge .............. 15.22  |==


OpenVINO 2023.1
Model: Vehicle Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 2043.05 |=======================
t2d-standard-60 AMD Milan . 1512.57 |=================
c6g.16xlarge .............. 0.14    |
m7a.16xlarge .............. 3666.89 |==========================================


OpenVINO 2023.1
Model: Vehicle Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 5.86    |
t2d-standard-60 AMD Milan . 9.90    |
c6g.16xlarge .............. 6990.10 |==========================================
m7a.16xlarge .............. 4.35    |


OpenVINO 2023.1
Model: Weld Porosity Detection FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 1875.28 |=========================
t2d-standard-60 AMD Milan . 1014.70 |==============
c6g.16xlarge .............. 8.39    |
m7a.16xlarge .............. 3132.48 |==========================================


OpenVINO 2023.1
Model: Weld Porosity Detection FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 15.98  |======
t2d-standard-60 AMD Milan . 14.76  |=====
c6g.16xlarge .............. 119.17 |===========================================
m7a.16xlarge .............. 10.19  |====


OpenVINO 2023.1
Model: Face Detection Retail FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 6605.12  |==========================
t2d-standard-60 AMD Milan . 4239.52  |=================
c6g.16xlarge .............. 0.46     |
m7a.16xlarge .............. 10382.22 |=========================================


OpenVINO 2023.1
Model: Face Detection Retail FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 4.53    |
t2d-standard-60 AMD Milan . 3.52    |
c6g.16xlarge .............. 2186.81 |==========================================
m7a.16xlarge .............. 3.07    |


OpenVINO 2023.1
Model: Road Segmentation ADAS FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 645.74  |======================
t2d-standard-60 AMD Milan . 565.34  |===================
c6g.16xlarge .............. 0.15    |
m7a.16xlarge .............. 1222.99 |==========================================


OpenVINO 2023.1
Model: Road Segmentation ADAS FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 18.56   |
t2d-standard-60 AMD Milan . 26.50   |
c6g.16xlarge .............. 6773.31 |==========================================
m7a.16xlarge .............. 13.07   |


OpenVINO 2023.1
Model: Machine Translation EN To DE FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 185.29 |=========================
t2d-standard-60 AMD Milan . 96.58  |=============
c6g.16xlarge .............. 1.36   |
m7a.16xlarge .............. 315.14 |===========================================


OpenVINO 2023.1
Model: Machine Translation EN To DE FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 64.70  |====
t2d-standard-60 AMD Milan . 155.14 |=========
c6g.16xlarge .............. 735.58 |===========================================
m7a.16xlarge .............. 50.72  |===


OpenVINO 2023.1
Model: Weld Porosity Detection FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 3650.57 |=========================
t2d-standard-60 AMD Milan . 2646.58 |==================
c6g.16xlarge .............. 5.50    |
m7a.16xlarge .............. 6177.94 |==========================================


OpenVINO 2023.1
Model: Weld Porosity Detection FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 8.20   |==
t2d-standard-60 AMD Milan . 11.32  |===
c6g.16xlarge .............. 181.94 |===========================================
m7a.16xlarge .............. 5.16   |=


OpenVINO 2023.1
Model: Person Vehicle Bike Detection FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 1764.21 |========================
t2d-standard-60 AMD Milan . 633.76  |========
c6g.16xlarge .............. 7.36    |
m7a.16xlarge .............. 3146.04 |==========================================


OpenVINO 2023.1
Model: Person Vehicle Bike Detection FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 6.79   |==
t2d-standard-60 AMD Milan . 23.64  |=======
c6g.16xlarge .............. 135.87 |===========================================
m7a.16xlarge .............. 5.07   |==


OpenVINO 2023.1
Model: Handwritten English Recognition FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 964.46  |=============================
t2d-standard-60 AMD Milan . 370.03  |===========
c6g.16xlarge .............. 2.53    |
m7a.16xlarge .............. 1419.11 |==========================================


OpenVINO 2023.1
Model: Handwritten English Recognition FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 31.08  |===
t2d-standard-60 AMD Milan . 81.00  |=========
c6g.16xlarge .............. 394.94 |===========================================
m7a.16xlarge .............. 22.53  |==


OpenVINO 2023.1
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 43607.04 |======================
t2d-standard-60 AMD Milan . 29668.12 |===============
c6g.16xlarge .............. 178.82   |
m7a.16xlarge .............. 81996.81 |=========================================


OpenVINO 2023.1
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 0.52 |====
t2d-standard-60 AMD Milan . 0.99 |========
c6g.16xlarge .............. 5.58 |=============================================
m7a.16xlarge .............. 0.38 |===


OpenVINO 2023.1
Model: Handwritten English Recognition FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 761.14  |============================
t2d-standard-60 AMD Milan . 390.72  |==============
c6g.16xlarge .............. 2.36    |
m7a.16xlarge .............. 1158.43 |==========================================


OpenVINO 2023.1
Model: Handwritten English Recognition FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 39.38  |====
t2d-standard-60 AMD Milan . 76.72  |========
c6g.16xlarge .............. 423.95 |===========================================
m7a.16xlarge .............. 27.60  |===


OpenVINO 2023.1
Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU
FPS > Higher Is Better
c3d-standard-60 AMD Genoa . 54971.26 |========================
t2d-standard-60 AMD Milan . 44049.15 |====================
c6g.16xlarge .............. 136.16   |
m7a.16xlarge .............. 92100.80 |=========================================


OpenVINO 2023.1
Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU
ms < Lower Is Better
c3d-standard-60 AMD Genoa . 0.40 |==
t2d-standard-60 AMD Milan . 0.61 |====
c6g.16xlarge .............. 7.33 |=============================================
m7a.16xlarge .............. 0.27 |==


Apache Cassandra 4.1.3
Test: Writes
Op/s > Higher Is Better
c3d-standard-60 AMD Genoa . 228640 |===================================
t2d-standard-60 AMD Milan . 187169 |=============================
c6g.16xlarge .............. 217355 |==================================
m7a.16xlarge .............. 278585 |===========================================


nginx 1.23.2
Connections: 500
Requests Per Second > Higher Is Better
c3d-standard-60 AMD Genoa . 187350.44 |================================
t2d-standard-60 AMD Milan . 162957.75 |============================
c6g.16xlarge .............. 162553.85 |============================
m7a.16xlarge .............. 233014.72 |========================================


nginx 1.23.2
Connections: 1000
Requests Per Second > Higher Is Better
c3d-standard-60 AMD Genoa . 180537.84 |================================
t2d-standard-60 AMD Milan . 155609.04 |============================
c6g.16xlarge .............. 158700.36 |============================
m7a.16xlarge .............. 224859.09 |========================================


BRL-CAD 7.36
VGR Performance Metric
VGR Performance Metric > Higher Is Better
c3d-standard-60 AMD Genoa . 510819 |============================
t2d-standard-60 AMD Milan . 629363 |==================================
m7a.16xlarge .............. 788704 |===========================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency
ms < Lower Is Better
t2d-standard-60 AMD Milan . 0.399 |=======================
c6g.16xlarge .............. 0.767 |============================================
m7a.16xlarge .............. 0.274 |================


PostgreSQL 16
Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency
ms < Lower Is Better
t2d-standard-60 AMD Milan . 0.498 |=====================
c6g.16xlarge .............. 1.026 |============================================
m7a.16xlarge .............. 0.347 |===============


PostgreSQL 16
Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency
ms < Lower Is Better
t2d-standard-60 AMD Milan . 140.92 |====================================
c6g.16xlarge .............. 168.19 |===========================================
m7a.16xlarge .............. 150.60 |=======================================


PostgreSQL 16
Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency
ms < Lower Is Better
t2d-standard-60 AMD Milan . 172.72 |===================================
c6g.16xlarge .............. 210.61 |===========================================
m7a.16xlarge .............. 188.68 |=======================================