AMD EPYC 7601 2P Ubuntu 19.10
2 x AMD EPYC 7601 32-Core testing with a Dell 02MJ3T (1.2.5 BIOS) and llvmpipe on Ubuntu 19.10 via the Phoronix Test Suite.


EPYC 7601 2P: 

	Processor: 2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD 17h, Memory: 504GB, Disk: 280GB INTEL SSDPED1D280GA + 12 x 500GB Samsung SSD 860 + 120GB INTEL SSDSCKJB120G7R, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA + 2 x Broadcom NetXtreme BCM5720 2-port PCIe

	OS: Ubuntu 19.10, Kernel: 5.5.0-rc7-phx-k10temp7 (x86_64) 20200123, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.8 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 1600x1200

Linux 5.8.10: 

	Processor: 2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD 17h, Memory: 504GB, Disk: 280GB INTEL SSDPED1D280GA + 12 x 500GB Samsung SSD 860 + 120GB INTEL SSDSCKJB120G7R, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA + 2 x Broadcom NetXtreme BCM5720 2-port PCIe

	OS: Ubuntu 19.10, Kernel: 5.8.10-050810-generic (x86_64), Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.8 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 1600x1200

Linux 5.9 Git: 

	Processor: 2 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads), Motherboard: Dell 02MJ3T (1.2.5 BIOS), Chipset: AMD 17h, Memory: 504GB, Disk: 280GB INTEL SSDPED1D280GA + 12 x 500GB Samsung SSD 860 + 120GB INTEL SSDSCKJB120G7R, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA + 2 x Broadcom NetXtreme BCM5720 2-port PCIe

	OS: Ubuntu 19.10, Kernel: 5.9.0-050900rc6daily20200922-generic (x86_64) 20200921, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: modesetting 1.20.5, OpenGL: 3.3 Mesa 19.2.8 (LLVM 9.0 128 bits), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 1600x1200


Timed LLVM Compilation 10.0
Time To Compile
Seconds < Lower Is Better
EPYC 7601 2P .. 246.98 |=============================
Linux 5.8.10 .. 268.16 |================================
Linux 5.9 Git . 461.08 |=======================================================


TensorFlow Lite 2020-08-23
Model: NASNet Mobile
Microseconds < Lower Is Better
EPYC 7601 2P .. 231136 |============================================
Linux 5.8.10 .. 230612 |============================================
Linux 5.9 Git . 286879 |=======================================================


perf-bench
Benchmark: Sched Pipe
ops/sec > Higher Is Better
EPYC 7601 2P .. 264219 |=======================================================
Linux 5.8.10 .. 226088 |===============================================
Linux 5.9 Git . 239236 |==================================================


AOM AV1 2.0
Encoder Mode: Speed 6 Realtime
Frames Per Second > Higher Is Better
EPYC 7601 2P .. 13.36 |========================================================
Linux 5.8.10 .. 12.21 |===================================================
Linux 5.9 Git . 12.61 |=====================================================


GROMACS 2020.1
Water Benchmark
Ns Per Day > Higher Is Better
EPYC 7601 2P .. 4.154 |========================================================
Linux 5.8.10 .. 4.002 |======================================================
Linux 5.9 Git . 3.886 |====================================================


TensorFlow Lite 2020-08-23
Model: Inception ResNet V2
Microseconds < Lower Is Better
EPYC 7601 2P .. 1033243 |====================================================
Linux 5.8.10 .. 999872  |===================================================
Linux 5.9 Git . 1067793 |======================================================


perf-bench
Benchmark: Epoll Wait
ops/sec > Higher Is Better
EPYC 7601 2P .. 1789 |=========================================================
Linux 5.8.10 .. 1682 |======================================================
Linux 5.9 Git . 1681 |======================================================


perf-bench
Benchmark: Syscall Basic
ops/sec > Higher Is Better
EPYC 7601 2P .. 14984082 |=====================================================
Linux 5.8.10 .. 14574671 |====================================================
Linux 5.9 Git . 14114717 |==================================================


GPAW 20.1
Input: Carbon Nanotube
Seconds < Lower Is Better
EPYC 7601 2P .. 83.94 |=====================================================
Linux 5.8.10 .. 89.04 |========================================================
Linux 5.9 Git . 85.15 |======================================================


AOM AV1 2.0
Encoder Mode: Speed 0 Two-Pass
Frames Per Second > Higher Is Better
EPYC 7601 2P .. 0.18 |=========================================================
Linux 5.8.10 .. 0.18 |=========================================================
Linux 5.9 Git . 0.17 |======================================================


TensorFlow Lite 2020-08-23
Model: Inception V4
Microseconds < Lower Is Better
EPYC 7601 2P .. 1079673 |====================================================
Linux 5.8.10 .. 1067050 |===================================================
Linux 5.9 Git . 1128075 |======================================================


AOM AV1 2.0
Encoder Mode: Speed 4 Two-Pass
Frames Per Second > Higher Is Better
EPYC 7601 2P .. 1.49 |=========================================================
Linux 5.8.10 .. 1.43 |=======================================================
Linux 5.9 Git . 1.41 |======================================================


Mobile Neural Network 2020-09-17
Model: SqueezeNetV1.0
ms < Lower Is Better
EPYC 7601 2P .. 17.10 |=======================================================
Linux 5.8.10 .. 16.52 |=====================================================
Linux 5.9 Git . 17.41 |========================================================


Incompact3D 2020-09-17
Input: Cylinder
Seconds < Lower Is Better
EPYC 7601 2P .. 228.35 |====================================================
Linux 5.8.10 .. 240.60 |=======================================================
Linux 5.9 Git . 234.59 |======================================================


perf-bench
Benchmark: Futex Lock-Pi
ops/sec > Higher Is Better
EPYC 7601 2P .. 81 |===========================================================
Linux 5.8.10 .. 77 |========================================================
Linux 5.9 Git . 79 |==========================================================


AOM AV1 2.0
Encoder Mode: Speed 8 Realtime
Frames Per Second > Higher Is Better
EPYC 7601 2P .. 22.28 |======================================================
Linux 5.8.10 .. 23.21 |========================================================
Linux 5.9 Git . 23.19 |========================================================


Blender 2.90
Blend File: Barbershop - Compute: CPU-Only
Seconds < Lower Is Better
EPYC 7601 2P .. 207.38 |=====================================================
Linux 5.8.10 .. 214.68 |=======================================================
Linux 5.9 Git . 207.19 |=====================================================


AOM AV1 2.0
Encoder Mode: Speed 6 Two-Pass
Frames Per Second > Higher Is Better
EPYC 7601 2P .. 2.38 |=========================================================
Linux 5.8.10 .. 2.32 |========================================================
Linux 5.9 Git . 2.33 |========================================================


ASTC Encoder 2.0
Preset: Thorough
Seconds < Lower Is Better
EPYC 7601 2P .. 6.32 |========================================================
Linux 5.8.10 .. 6.48 |=========================================================
Linux 5.9 Git . 6.34 |========================================================


Zstd Compression 1.4.5
Compression Level: 3
MB/s > Higher Is Better
EPYC 7601 2P .. 4574.5 |=======================================================
Linux 5.8.10 .. 4572.9 |=======================================================
Linux 5.9 Git . 4466.6 |======================================================


Monte Carlo Simulations of Ionised Nebulae 2019-03-24
Input: Dust 2D tau100.0
Seconds < Lower Is Better
EPYC 7601 2P .. 261 |==========================================================
Linux 5.8.10 .. 255 |=========================================================
Linux 5.9 Git . 255 |=========================================================


Mobile Neural Network 2020-09-17
Model: inception-v3
ms < Lower Is Better
EPYC 7601 2P .. 61.09 |========================================================
Linux 5.8.10 .. 59.70 |=======================================================
Linux 5.9 Git . 60.82 |========================================================


Blender 2.90
Blend File: Pabellon Barcelona - Compute: CPU-Only
Seconds < Lower Is Better
EPYC 7601 2P .. 134.45 |======================================================
Linux 5.8.10 .. 135.57 |======================================================
Linux 5.9 Git . 136.86 |=======================================================


NAMD 2.14
ATPase Simulation - 327,506 Atoms
days/ns < Lower Is Better
EPYC 7601 2P .. 0.46809 |=====================================================
Linux 5.8.10 .. 0.46431 |=====================================================
Linux 5.9 Git . 0.47248 |======================================================


ASTC Encoder 2.0
Preset: Medium
Seconds < Lower Is Better
EPYC 7601 2P .. 7.50 |========================================================
Linux 5.8.10 .. 7.62 |=========================================================
Linux 5.9 Git . 7.49 |========================================================


Mobile Neural Network 2020-09-17
Model: mobilenet-v1-1.0
ms < Lower Is Better
EPYC 7601 2P .. 12.49 |========================================================
Linux 5.8.10 .. 12.29 |=======================================================
Linux 5.9 Git . 12.30 |=======================================================


Mobile Neural Network 2020-09-17
Model: resnet-v2-50
ms < Lower Is Better
EPYC 7601 2P .. 64.97 |=======================================================
Linux 5.8.10 .. 66.01 |========================================================
Linux 5.9 Git . 65.09 |=======================================================


LAMMPS Molecular Dynamics Simulator 24Aug2020
Model: Rhodopsin Protein
ns/day > Higher Is Better
EPYC 7601 2P .. 24.52 |========================================================
Linux 5.8.10 .. 24.18 |=======================================================
Linux 5.9 Git . 24.24 |=======================================================


Blender 2.90
Blend File: Classroom - Compute: CPU-Only
Seconds < Lower Is Better
EPYC 7601 2P .. 104.13 |=======================================================
Linux 5.8.10 .. 104.79 |=======================================================
Linux 5.9 Git . 104.99 |=======================================================


System GZIP Decompression

Seconds < Lower Is Better
EPYC 7601 2P .. 3.855 |========================================================
Linux 5.8.10 .. 3.882 |========================================================
Linux 5.9 Git . 3.884 |========================================================


perf-bench
Benchmark: Futex Hash
ops/sec > Higher Is Better
EPYC 7601 2P .. 1930718 |======================================================
Linux 5.8.10 .. 1942755 |======================================================
Linux 5.9 Git . 1932178 |======================================================


Blender 2.90
Blend File: Fishy Cat - Compute: CPU-Only
Seconds < Lower Is Better
EPYC 7601 2P .. 68.61 |========================================================
Linux 5.8.10 .. 68.56 |========================================================
Linux 5.9 Git . 68.88 |========================================================


ASTC Encoder 2.0
Preset: Fast
Seconds < Lower Is Better
EPYC 7601 2P .. 6.63 |=========================================================
Linux 5.8.10 .. 6.60 |=========================================================
Linux 5.9 Git . 6.63 |=========================================================


LAMMPS Molecular Dynamics Simulator 24Aug2020
Model: 20k Atoms
ns/day > Higher Is Better
EPYC 7601 2P .. 24.77 |========================================================
Linux 5.8.10 .. 24.74 |========================================================
Linux 5.9 Git . 24.71 |========================================================


ASTC Encoder 2.0
Preset: Exhaustive
Seconds < Lower Is Better
EPYC 7601 2P .. 44.93 |========================================================
Linux 5.8.10 .. 45.03 |========================================================
Linux 5.9 Git . 45.04 |========================================================


Blender 2.90
Blend File: BMW27 - Compute: CPU-Only
Seconds < Lower Is Better
EPYC 7601 2P .. 42.58 |========================================================
Linux 5.8.10 .. 42.52 |========================================================
Linux 5.9 Git . 42.48 |========================================================


Kripke 1.2.4

Throughput FoM > Higher Is Better
EPYC 7601 2P .. 34516849 |=====================================================
Linux 5.8.10 .. 29824966 |==============================================
Linux 5.9 Git . 30386885 |===============================================


NCNN 20200916
Target: CPU - Model: mobilenetv2_yolov3
ms < Lower Is Better
EPYC 7601 2P .. 75.84  |===============
Linux 5.8.10 .. 67.16  |=============
Linux 5.9 Git . 283.78 |=======================================================


NCNN 20200916
Target: CPU - Model: resnet50_int8
ms < Lower Is Better
EPYC 7601 2P .. 204.62 |=========================
Linux 5.8.10 .. 168.54 |=====================
Linux 5.9 Git . 450.58 |=======================================================


NCNN 20200916
Target: CPU - Model: alexnet
ms < Lower Is Better
EPYC 7601 2P .. 62.36  |==============================
Linux 5.8.10 .. 59.29  |============================
Linux 5.9 Git . 115.69 |=======================================================


NCNN 20200916
Target: CPU - Model: resnet18_int8
ms < Lower Is Better
EPYC 7601 2P .. 72.59  |=================
Linux 5.8.10 .. 64.51  |===============
Linux 5.9 Git . 241.57 |=======================================================


NCNN 20200916
Target: CPU - Model: vgg16_int8
ms < Lower Is Better
EPYC 7601 2P .. 188.28 |=========================================
Linux 5.8.10 .. 155.97 |==================================
Linux 5.9 Git . 252.48 |=======================================================


NCNN 20200916
Target: CPU - Model: googlenet_int8
ms < Lower Is Better
EPYC 7601 2P .. 171.39 |====================
Linux 5.8.10 .. 147.63 |=================
Linux 5.9 Git . 467.54 |=======================================================


NCNN 20200916
Target: CPU - Model: blazeface
ms < Lower Is Better
EPYC 7601 2P .. 19.64 |=================
Linux 5.8.10 .. 17.79 |================
Linux 5.9 Git . 63.91 |========================================================


NCNN 20200916
Target: CPU - Model: mnasnet
ms < Lower Is Better
EPYC 7601 2P .. 39.71 |========================
Linux 5.8.10 .. 37.53 |=======================
Linux 5.9 Git . 91.63 |========================================================


NCNN 20200916
Target: CPU - Model: squeezenet
ms < Lower Is Better
EPYC 7601 2P .. 48.38 |================================
Linux 5.8.10 .. 41.95 |============================
Linux 5.9 Git . 84.53 |========================================================


NCNN 20200916
Target: CPU - Model: mobilenet_v3
ms < Lower Is Better
EPYC 7601 2P .. 40.64 |========================
Linux 5.8.10 .. 46.30 |===========================
Linux 5.9 Git . 96.67 |========================================================


NCNN 20200916
Target: CPU - Model: squeezenet_int8
ms < Lower Is Better
EPYC 7601 2P .. 58.76 |===================================
Linux 5.8.10 .. 54.92 |=================================
Linux 5.9 Git . 94.26 |========================================================


Mobile Neural Network 2020-09-17
Model: MobileNetV2_224
ms < Lower Is Better
EPYC 7601 2P .. 10.50 |=======================================================
Linux 5.8.10 .. 10.10 |=====================================================
Linux 5.9 Git . 10.73 |========================================================


TensorFlow Lite 2020-08-23
Model: Mobilenet Quant
Microseconds < Lower Is Better
EPYC 7601 2P .. 51765.5 |=============================================
Linux 5.8.10 .. 51422.6 |=============================================
Linux 5.9 Git . 62210.3 |======================================================


TensorFlow Lite 2020-08-23
Model: Mobilenet Float
Microseconds < Lower Is Better
EPYC 7601 2P .. 51413.1 |============================================
Linux 5.8.10 .. 49873.7 |==========================================
Linux 5.9 Git . 63482.5 |======================================================


TensorFlow Lite 2020-08-23
Model: SqueezeNet
Microseconds < Lower Is Better
EPYC 7601 2P .. 90123.7  |=============================================
Linux 5.8.10 .. 87243.7  |===========================================
Linux 5.9 Git . 106374.2 |=====================================================


perf-bench
Benchmark: Memset 1MB
GB/sec > Higher Is Better
EPYC 7601 2P .. 41.37 |========================================================
Linux 5.8.10 .. 33.65 |==============================================
Linux 5.9 Git . 38.08 |====================================================


perf-bench
Benchmark: Memcpy 1MB
GB/sec > Higher Is Better
EPYC 7601 2P .. 14.56 |========================================================
Linux 5.8.10 .. 13.50 |====================================================
Linux 5.9 Git . 14.49 |========================================================


Zstd Compression 1.4.5
Compression Level: 19
MB/s > Higher Is Better
EPYC 7601 2P .. 76.2 |=========================================================
Linux 5.8.10 .. 72.7 |======================================================
Linux 5.9 Git . 73.8 |=======================================================