slurm

name	slurm
description	Submit, monitor, and manage SLURM compute jobs on the PVE cluster (ctl + node01). Covers sbatch, srun, squeue, scancel, sacct, and MiQi-specific job templates.
metadata	{"miqi":{"emoji":"⚙️","os":["linux"],"notes":"Windows users connect via SSH"}}

SLURM Skill

Use this skill when the user wants to submit, monitor, cancel, or inspect jobs on the SLURM cluster.

Cluster & Access

Host	Role	Access
ctl	Control node — submit jobs	ssh ctl

Windows 用户: SLURM 命令无法在本地运行，必须先 ssh ctl，再在 ctl 上执行。也可 ssh ctl "squeue" 直接远程执行。

上传脚本: scp job.sh ctl:~ | 取回日志: scp "ctl:~/slurm-%j.out" .

Quick Reference

sinfo                         # cluster/partition status
squeue -u $USER               # my jobs
sbatch job.sh                 # submit
srun --pty bash               # interactive shell
scancel <jobid>               # cancel job
scancel -u $USER              # cancel all my jobs
sacct -j <jobid> --format=JobID,State,Elapsed,MaxRSS
scontrol show job <jobid>     # job details

Windows Quick Reference

ssh ctl "sbatch ~/job.sh" # submit 不要使用转义 按照示例格式输入
ssh ctl "squeue -u $USER"  # my jobs
ssh ctl "scancel <jobid>"  # cancel job
ssh ctl "sacct -j <jobid>" # job details

不需要把脚本从 Windows 格式转成 Linux 格式

查看任务输出

目前没有nfs共享文件系统，所以不能查看任务输出