search

fengggli / gpu-computing-materials

A simple deep learning framework that optimizes task scheduling and memory usage on different CPU/GPU architectures.
1 stars 0 forks source link

server information #58

Open fengggli opened 5 years ago

fengggli commented 5 years ago

This page records server configurations.

fengggli commented 5 years ago
server name cpu memory OS link
sievert 2 x Intel(R) Xeon(R) CPU E5-2640 0 @ 2.50GHz, each has 6 cores and x2 hyperthreading, with avx 64GB ubuntu 16.04 https://lenovopress.com/tips0850-system-x3650-m4-e5-2600-v2
stampede2 skx 2 x Intel Xeon Platinum 8160, )each has 24 cores and x2 hyperthreading, with avx, avx2, avx512 192GB (2.67GHz) DDR4 RHEL 7 specs from tacc
Gibson 2x Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz, each has 20 cores and x2 hyperthreading, with avx, avx2, avx512 128G RHEL 7.7 cpu info from intel
rain/thuner/lightning 2 x Intel Xeon Xeon E5-2630v3, )each has 8 cores and x2 hyperthreading, with avx, avx2 32G some Linux ask Debbie
fengggli commented 4 years ago

bigred iii

  1. Check CLE version 
    lifen@elogin2:~> cat /etc/opt/cray/release/cle-release
RELEASE=6.0.UP06
BUILD=6.0.6734
DATE=20190919
ARCH=noarch
NETWORK=ari
PATCHSET=41-201909191725

UP06 datawarp