Re: [slurm-users] CPU allocation within a node is not cyclic

2018-10-07 Thread CUI Hao
Changing SelectTypeParameters to CR_Core_Memory still does not guarantee cyclic core allocation. I dig into Slurm source code and find another potential issues in select/cons_res plugin. For detail: https://bugs.schedmd.com/show_bug.cgi?id=5819 崔灏 (CUI Hao) 于2018年10月6日周六 上午10:15写道: >

Re: [slurm-users] CPU allocation within a node is not cyclic

2018-10-05 Thread CUI Hao
restarting slurmctld will interrupt current tasks, so I'm still waiting for them to finish. 崔灏 (CUI Hao) 于2018年10月5日周五 下午10:12写道: > > According to https://slurm.schedmd.com/cpu_management.html, > > The default allocation method within a node is cyclic allocation (allocate > >

[slurm-users] CPU allocation within a node is not cyclic

2018-10-05 Thread CUI Hao
560c5d8228a I really hope to get the default cylic allocation behaviour, so that CPU jobs won't take up cores assigned to GPU devices. Any idea what's wrong with our configs? -- 崔灏 / CUI Hao Homepage: i-yu.me Twitter: @cuihaoleo

[slurm-users] Gres CPU binding in 17.02.10

2018-05-30 Thread CUI Hao
f it is a bug hasn't been fixed in 17.02.10, or something went wrong with our configuration? Here is our gres.conf: NodeName=wmc-slave-g[1-3] Name=gpu File=/dev/nvidia[0-1] CPUs=0-11 NodeName=wmc-slave-g[1-3] Name=gpu File=/dev/nvidia[2-3] CPUs=12-23 [1] https://bugs.schedmd.com/show_bug.cgi?