Olm’s “pestat” script does allow you to get similar information, but I’m
interested to see if indeed there’s a better answer. I’ve used his script for
more or less the same reason, to see if the jobs are using the resources
they’re allocated. They show at a node level though, and then you have to look
closer. For example:
Print only nodes that are flagged by * (RED nodes)
Hostname Partition Node Num_CPU CPUload Memsize Freemem Joblist
State Use/Tot (MB) (MB) JobId
User ...
gpu003 oarc drng* 8 12 58.06* 64000 24507 82565618
yc567
...
hal0027 kopp_1 alloc 28 28 8.64* 128000 115610 82591085
mes373 82595703 aek119
You can see, both of the above are examples of jobs that have allocated CPU
numbers that are very different from the ultimate CPU load (the first one using
way more than allocated, though they’re in a cgroup so theoretically isolated
from the other users on the machine), and the second one asking for all 28 CPUs
but only “using” ~8 of them.
If you’re using cgroups, it would seem to me that there must also be a way to
see the output of “top” for just a group, or at least something similar.
systemd-cgtop does more or less that, but doesn’t seem to show exactly what
you’d want here:
Path
Tasks %CPU Memory Input/s Output/s
/
306 900.6 9.8G - -
/slurm
- - 3.7G - -
/slurm/uid_140780
- - 3.0G - -
/slurm/uid_140780/job_82591085
- - 3.0G - -
/slurm/uid_142473
- - 374.7M - -
/slurm/uid_142473/job_82595703
- - 374.7M - -
…CPU only being shown as an aggregate at the top level (sorry about the
formatting).
> On Nov 21, 2018, at 1:27 PM, 宋亚磊 <[email protected]> wrote:
>
> Hi Jing, thank you!
>
> The following command show us the cpu load of the node,
>
> $ scontrol show node <nodename_of_job> | grep CPULoad
>
> but I want the percent cpu of the job, like top or ps.
> For examplt, a job allocated 10 cpus, but it just use 2, so the percent
> cpu should be 200%, not be 1000%, I want konw this.
>
> Anyway, thank you again, Jing.
>
> Best regards,
> Yalei
>
>> -----原始邮件-----
>> 发件人: "Jing Gong" <[email protected]>
>> 发送时间: 2018-11-22 02:04:59 (星期四)
>> 收件人: "Slurm User Community List" <[email protected]>
>> 抄送:
>> 主题: Re: [slurm-users] How to check the percent cpu of a job?
>>
>> Hi,
>>
>>> How to check the percent cpu of a job in slurm?
>>
>> We use command "scontrol" likes
>>
>> $ scontrol show node <nodename_of_job> | grep CPULoad
>> ...
>> CPUAlloc=48 CPUErr=0 CPUTot=48 CPULoad=25.32
>> ...
>>
>> Regards, Jing
>>
>> ________________________________________
>> From: slurm-users <[email protected]> on behalf of 宋亚磊
>> <[email protected]>
>> Sent: Wednesday, November 21, 2018 18:51
>> To: [email protected]
>> Subject: [slurm-users] How to check the percent cpu of a job?
>>
>> Hello everyone,
>>
>> How to check the percent cpu of a job in slurm? I tried sacct, sstat,
>> squeue, but I can't find that how to check.
>> Can someone help me?
>>
>> Best regards,
>> Yalei
>>
--
____
|| \\UTGERS, |---------------------------*O*---------------------------
||_// the State | Ryan Novosielski - [email protected]
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'
