[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [ptp-user] System Monitoring && Generic Slurm Batch
|
Hi Christoph,
You are welcome to update the documentation if you see any errors.
Regarding the naming, is this the name of the directory under “rms” or the name of the target configuration (from the xml file)? If the former, then the current ones are SLURM and SLURM_ALPS. I’d suggest renaming SLURM to SLURM_BG (if it is really BG specific) and using SLURM_INTEL for yours.
Regards,
Greg
> On Oct 14, 2015, at 3:04 AM, Christoph Pospiech <cpospiech@xxxxxxxxxx> wrote:
>
> On Monday, October 12, 2015 12:00:16 PM I wrote:
>> Hi,
>>
>> I have tried to use the above configuration, but the main display
>> doesn't show the nodes correctly. What is needed to make that work ?
>> How can I debug this issue ?
>>
>> BTW, it works with the same Eclipse installation for other remote
>> hosts that use different batch schedulers, such as LSF.
>
> Hi,
>
> I made some progress in this matter. Hopefully I can now ask
> more specific questions.
>
> Somehow I stumbled over the magic letters LML (DA), and
> that rang a bell with me. I sent uncle Google hunting and he
> came back with https://wiki.eclipse.org/PTP/LML_DA_Driver .
>
> On this page I found the broken link
> http://llview.zam.kfa-juelich.de/LML/OnlineDocumentation/lmldoc.html
> Fortunately I am sufficiently familiar with FZ Juelich matters to know
> that any "zam.kfa-juelich" should better be replaced by "fz-juelich".
> That worked here as well. Can I/Should I fix it myself on wiki,eclipse.org ?
> (I don't want to meddle with documents that others may regard as theirs).
>
> I guess I found the perl LML script responsible for the broken display above.
> Apparently it is not totally generic, but written with Blue Gene in mind.
> So it has to be adapted to Intel clusters. Not sure whether I have the time
> to amend this, but suppose so and suppose it turns out that there cannot be
> an one-size-fits-all LML DA driver for both Intel cluster and BlueGene.
> What naming convention should I follow ? Leave "SLURM generic" to
> BlueGene and put a "SLURM cluster generic" next to it ? Or rename
> 'SLURM generic' to 'SLURM BlueGene' - which is a disruptive change -
> and keep the name 'SLURM generic' for Linux clusters ?
>
> Comments welcome !
> --
>
> Mit freundlichen Grüßen / Kind regards
>
> Dr. Christoph Pospiech
> High Performance & Parallel Computing
> Phone: +49-351 86269826
> Mobile: +49-171-765 5871
> E-Mail: cpospiech@xxxxxxxxxx
>
> Lenovo (Deutschland) GmbH
> Meitnerstr. 9
> D-70563 Stuttgart
>
> Geschäftsführung: Bernhard Fauser
> Sitz der Gesellschaft: Stuttgart
> HRB-Nr.: 25189, AG Stuttgart
> WEEE-Reg.-Nr.: DE79679404
> _______________________________________________
> ptp-user mailing list
> ptp-user@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/ptp-user
