parallel-processing - slurm controller busy a lot of times when used by other users

Question

I am fairly new to SLURM: the grid I use has many different users and when they are submitting or canceling jobs, it seems that other users are not able to query partition status, etc. This is extremely frustrating especially when creating jobs that spawn other jobs since they end up failing because the controller is busy. Does anyone know a workaround?

Answer 1

在默认设置下，当许多用户同时提交/修改/取消许多作业时，Slurm 可能会变慢/挂起，尤其是在启用回填和记帐的情况下。

请参阅2012 年 Slurm 用户组会议的这些幻灯片中的改进提示。