[Clusterusers] the cluster room must be really hot right now
Bassam Kurdali
bassam at urchn.org
Wed Jun 19 13:37:30 EDT 2013
Cool, I was just checking to see if it was normal- in our case if a job
took that long we'd think there was a problem with it. I got curious
when none of our renders spooled on the cluster (they're running on the
macs) and clicked on the nodes in tractor to find out.
cluster top sounds neat though, I hope they port it to their next
release.
On Wed, 2013-06-19 at 09:07 -0400, Wm. Josiah Erikson wrote:
> No, I don't think it's bad. The way to tell what's going on on what
> node is to use the tractor interface or to log into the nodes. I miss
> cluster top too. I asked the ROCKS developers about it, and they said
> they didn't have time to port it to this release.
> -Josiah
>
>
> On 6/19/13 8:03 AM, Thomas Helmuth wrote:
>
> > A lot of what's running are my GP runs. Right now I have 101
> > heavy-duty tasks going on the cluster. I can tone it down a bit if
> > we think that's bad for some reason.
> >
> >
> > -Tom
> >
> >
> >
> > On Wed, Jun 19, 2013 at 7:54 AM, Lee Spector
> > <lspector at hampshire.edu> wrote:
> >
> > There used to be a "cluster top" link that would show what
> > "top" shows from a command line -- e.g. the specific
> > commands running with their runtimes. But I don't see it
> > now... Am I just missing it or did it go away? Hard to tell
> > what's running without it...
> >
> > -Lee
> >
> > On Jun 18, 2013, at 7:20 PM, Bassam Kurdali wrote:
> >
> > > Looking at the cluster right now I'm seeing many machines
> > with MaxLoad
> > > exceeded, and some jobs seem to be stuck on them for more
> > than a day, it
> > > could be ok (just heavy computation tasks) but just wanted
> > to poke in
> > > case something is going wrong.
> > > cheers
> > > Bassam
> > >
> > > On Wed, 2013-06-12 at 09:24 -0400, Wm. Josiah Erikson
> > wrote:
> > >> I was noting this as well. The best thing about it is
> > that between our
> > >> new ceiling-mounted 4-ton capacity Mitsubishi A/C unit
> > (thanks Mark
> > >> Monaco!) and our new energy-efficient Rack2, it's
> > actually quite cool in
> > >> there. Time to add more nodes, I guess :)
> > >>
> > >> -Josiah
> > >>
> > >>
> > >> On 6/11/13 11:45 PM, Jaime Davila wrote:
> > >>> Hey gang,
> > >>>
> > >>> If you haven't seen the ganglia report web page in the
> > last day, check
> > >>> it out. It's impressive, I've never seen it like that.
> > No one can say
> > >>> we don't take advantage of all that computing power!
> > >>>
> > >>> http://fly.hampshire.edu/ganglia/
> > >>>
> > >>
> > >
> > >
> > > _______________________________________________
> > > Clusterusers mailing list
> > > Clusterusers at lists.hampshire.edu
> > > https://lists.hampshire.edu/mailman/listinfo/clusterusers
> >
> >
> > --
> > Lee Spector, Professor of Computer Science
> > Cognitive Science, Hampshire College
> > 893 West Street, Amherst, MA 01002-3359
> > lspector at hampshire.edu, http://hampshire.edu/lspector/
> > Phone: 413-559-5352, Fax: 413-559-5438
> >
> > _______________________________________________
> > Clusterusers mailing list
> > Clusterusers at lists.hampshire.edu
> > https://lists.hampshire.edu/mailman/listinfo/clusterusers
> >
> >
> >
> >
> >
> > _______________________________________________
> > Clusterusers mailing list
> > Clusterusers at lists.hampshire.edu
> > https://lists.hampshire.edu/mailman/listinfo/clusterusers
>
> --
> Wm. Josiah Erikson
> Assistant Director of IT, Infrastructure Group
> System Administrator, School of CS
> Hampshire College
> Amherst, MA 01002
> (413) 559-6091
> _______________________________________________
> Clusterusers mailing list
> Clusterusers at lists.hampshire.edu
> https://lists.hampshire.edu/mailman/listinfo/clusterusers
More information about the Clusterusers
mailing list