I suspect that the scripts which OSCAR runs on detect the cpu count
but do not detect multiple cores as separate "processors".  If this is
the case, this is a fairly major issue and needs to be addressed soon.
 Multi-core processors are becoming more and more common.  I don't
have any hardware to test this on myself yet, but I should in a couple
months.

On the plus side, it ought to be a fairly easy fix...

On 22 Jan 2007 20:37:59 -0500, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
> Thanks for the message!
>
> Actually, I only have two client nodes. One have 4 dual-core AMD CPUs
> (total 8 logic CPUs). The other host has 2 dual-core CPUs (total 4 logic
> CPUs).
>
> I use "qsub ./script.sh" to submit jobs. All the jobs use the same script.
>
> If I submit 12 jobs, only 6 jobs runs (4 on the first node and 2 on the
> other node) at the same time.
>
> Here is what I get if I do "print server" from qmgr. Do I need to change
> anything?
>
> ======================================================
>
> #
> # Create queues and set their attributes.
> #
> #
> # Create and define queue workq
> #
> create queue workq
> set queue workq queue_type = Execution
> set queue workq resources_max.cput = 10000:00:00
> set queue workq resources_max.ncpus = 12
> set queue workq resources_max.nodect = 2
> set queue workq resources_max.walltime = 10000:00:00
> set queue workq resources_min.cput = 00:00:01
> set queue workq resources_min.ncpus = 1
> set queue workq resources_min.nodect = 1
> set queue workq resources_min.walltime = 00:00:01
> set queue workq resources_default.cput = 10000:00:00
> set queue workq resources_default.ncpus = 1
> set queue workq resources_default.nodect = 1
> set queue workq resources_default.walltime = 10000:00:00
> set queue workq resources_available.nodect = 2
> set queue workq enabled = True
> set queue workq started = True
> #
> # Set server attributes.
> #
> set server scheduling = True
> set server default_queue = workq
> set server log_events = 64
> set server mail_from = adm
> set server query_other_jobs = True
> set server resources_available.ncpus = 12
> set server resources_available.nodect = 2
> set server resources_available.nodes = 2
> set server resources_max.ncpus = 12
> set server resources_max.nodes = 2
> set server scheduler_iteration = 60
> set server node_check_rate = 150
> set server tcp_timeout = 6
> set server pbs_version = 2.0.0p8
>
> ===========================================================
>
>
> Here is maui.cfg file
>
> ===========================================================
>
> # maui.cfg 3.2.6p14
>
> SERVERHOST              photon.bwh.harvard.edu
> # primary admin must be first in list
> ADMIN1                root
>
> # Resource Manager Definition
>
> RMCFG[DUAL.EFOCHT.DE] TYPE=PBS
>
> # Allocation Manager Definition
>
> AMCFG[bank]  TYPE=NONE
>
> # full parameter docs at
> http://clusterresources.com/mauidocs/a.fparameters.html
> # use the 'schedctl -l' command to display current configuration
>
> RMPOLLINTERVAL  00:00:10
>
> SERVERPORT            42559
> SERVERMODE            NORMAL
>
> # Admin: http://clusterresources.com/mauidocs/a.esecurity.html
>
>
> LOGFILE               maui.log
> LOGFILEMAXSIZE        10000000
> LOGLEVEL              3
>
> # Job Priority:
> http://clusterresources.com/mauidocs/5.1jobprioritization.html
>
> QUEUETIMEWEIGHT       1
>
> # FairShare: http://clusterresources.com/mauidocs/6.3fairshare.html
>
> #FSPOLICY              PSDEDICATED
> #FSDEPTH               7
> #FSINTERVAL            86400
> #FSDECAY               0.80
>
> # Throttling Policies:
> http://clusterresources.com/mauidocs/6.2throttlingpolicies.html
>
> # NONE SPECIFIED
>
> # Backfill: http://clusterresources.com/mauidocs/8.2backfill.html
>
> BACKFILLPOLICY  ON
> RESERVATIONPOLICY     CURRENTHIGHEST
>
> # Node Allocation:
> http://clusterresources.com/mauidocs/5.2nodeallocation.html
>
> NODEALLOCATIONPOLICY  MINRESOURCE
>
> # QOS: http://clusterresources.com/mauidocs/7.3qos.html
>
> # QOSCFG[hi]  PRIORITY=100 XFTARGET=100 FLAGS=PREEMPTOR:IGNMAXJOB
> # QOSCFG[low] PRIORITY=-1000 FLAGS=PREEMPTEE
>
> # Standing Reservations:
> http://clusterresources.com/mauidocs/7.1.3standingreservations.html
>
> # SRSTARTTIME[test] 8:00:00
> # SRENDTIME[test]   17:00:00
> # SRDAYS[test]      MON TUE WED THU FRI
> # SRTASKCOUNT[test] 20
> # SRMAXTIME[test]   0:30:00
>
> # Creds: http://clusterresources.com/mauidocs/6.1fairnessoverview.html
>
> # USERCFG[DEFAULT]      FSTARGET=25.0
> # USERCFG[john]         PRIORITY=100  FSTARGET=10.0-
> # GROUPCFG[staff]       PRIORITY=1000 QLIST=hi:low QDEF=hi
> # CLASSCFG[batch]       FLAGS=PREEMPTEE
> # CLASSCFG[interactive] FLAGS=PREEMPTOR
>
> NODEACCESSPOLICY
>
> =====================================================================
>
>
>
> > Check ganglia at http://localhost/ganglia and see where those 6 jobs
> > are.  Make sure in particular they are not all sitting on one node or
> > something silly.  If you have 6 nodes and they are one per node, then
> > the queue is probably set up to reserve an entire node for each
> > process.  There is a flag in the torque config file (I think) that
> > tells it to do this.
> >
> > Could you post the script you are queueing with and the qsub command
> > you use to submit the job?
> >
> > Are you running an smp kernel on your head node, I assume?  If you
> > happened to be running in the non-smp version when you installed
> > torque/maui probably don't know that there is more than one processor
> > available...
> >
> > Hopefully this gives you some thoughts as to where to start looking...
> >
> > On 22 Jan 2007 13:13:02 -0500, Jinsong Ouyang <[EMAIL PROTECTED]>
> > wrote:
> >>
> >>
> >>
> >>
> >> I am using OSCAR 5.0 & Fedora 5.0 x86_64.  I have total 12 logic CPUs on
> >> computing nodes. I use qsub to submit jobs and can only have maximum 6
> >> jobs
> >> running simultaneously. Half of the CPUs are not used. Could anyone
> >> please
> >> tell me how to increase the number of running jobs? I tried to set
> >> max_running using qmgr. It does not seem to change anything. Do I need
> >> to
> >> change anything in maui.cfg?
> >>
> >>
> >>
> >> Many thanks,
> >>
> >>
> >>
> >> JO
> >> -------------------------------------------------------------------------
> >> Take Surveys. Earn Cash. Influence the Future of IT
> >> Join SourceForge.net's Techsay panel and you'll get the chance to share
> >> your
> >> opinions on IT & business topics through brief surveys - and earn cash
> >> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> >>
> >> _______________________________________________
> >> Oscar-users mailing list
> >> [email protected]
> >> https://lists.sourceforge.net/lists/listinfo/oscar-users
> >>
> >>
> >>
> >
> > -------------------------------------------------------------------------
> > Take Surveys. Earn Cash. Influence the Future of IT
> > Join SourceForge.net's Techsay panel and you'll get the chance to share
> > your
> > opinions on IT & business topics through brief surveys - and earn cash
> > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> > _______________________________________________
> > Oscar-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/oscar-users
> >
>
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Oscar-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/oscar-users
>

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Oscar-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oscar-users

Reply via email to