Slurm - Tag - Xiaopeng Xuhttps://xu-xp.com/tags/slurm/Slurm - Tag - Xiaopeng XuHugo -- gohugo.ioenxiaopeng.xu@kaust.edu.sa (Xiaopeng Xu)xiaopeng.xu@kaust.edu.sa (Xiaopeng Xu)Sat, 24 Oct 2020 00:00:00 +0000Slurm 常用命令https://xu-xp.com/posts/slurm_commons/Sat, 24 Oct 2020 00:00:00 +0000xiaopeng.xu@kaust.edu.sa (Xiaopeng Xu)https://xu-xp.com/posts/slurm_commons/相关培训

https://www.hpc.kaust.edu.sa/content/data-science-training

包括：

  1. Trillion-parameter scale model training and inference with DeepSpeed

    1. Codes: https://github.com/kaust-rccl/deepspeed_workshop/tree/master/HelloDeepSpeed, https://github.com/microsoft/Megatron-DeepSpeed

  2. Introduction to Containers on KSL Platforms

  3. High Throughput Hyperparameter Optimization on KSL platforms

  4. Distributed Deep Learning on KSL platforms

  5. Data Science on-boarding on KSL platforms

基础操作

登录 iBex

外网时，需要先连接 KAUST VPN。通常使用 Cisco anyconnect

登录 iBex：

]]>