Machine Learning Lab

User Tools

Site Tools

You are here: Laboratory for Neural Computing for Machine Learning (NCML) » Check CPUs workload:
Trace:
check_workload_cpu_gpu

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
check_workload_cpu_gpu [2017/07/10 20:05] – hjcheck_workload_cpu_gpu [2017/07/10 20:06] (current) – hj
Line 1: Line 1:
  === Check CPUs workload: ===  === Check CPUs workload: ===
-     - We have a simple script for you to check the workload of all machines, you may run:  **/cs/home/hj/bin/available_computers.pl**. Every time when you submit a new job, please use this command to look for a free or light-loaded machine. For a 6-core machine, we normally should not have its workload over 6. +  - We have a simple script for you to check the workload of all machines, you may run:  **/cs/home/hj/bin/available_computers.pl**. Every time when you submit a new job, please use this command to look for a free or light-loaded machine. For a 6-core machine, we normally should not have its workload over 6. 
-    - Run Linux '**htop**' command to check the CPU load, memory usage in each machines. +  - Run Linux '**htop**' command to check the CPU load, memory usage in each machines. 
-    - If your program consumes lots of memory (over 10G), DON’T submit it more than once to a single machine.+  - If your program consumes lots of memory (over 10G), DON’T submit it more than once to a single machine.
  
  === Check GPUs workload: ===  === Check GPUs workload: ===
check_workload_cpu_gpu.1499717136.txt.gz · Last modified: 2017/07/10 20:05 by hj