[GE users] Increasing slots past the number of CPUs?

kjpursley kevin.pursley at bp.com
Mon Jan 11 19:26:20 GMT 2010


For the entire queue modify the cluster queue with qmon, general
configuration tab.

Below is output from a queue with different sized machines.
Think slots is a integer do granularity is 1 slot.



qconf -sq comet

qname                 comet
hostlist              @comet
seq_no                0
load_thresholds       np_load_avg=1.75
suspend_thresholds    NONE
nsuspend              1
suspend_interval      00:05:00
priority              0
min_cpu_interval      00:05:00
processors            UNDEFINED
qtype                 BATCH INTERACTIVE
ckpt_list             NONE
pe_list               comet
rerun                 FALSE
slots                 64,[hpcal011=8],[hpca4201=8],[hpca4205=64], \
                      [hpca4206=64],[hpca4203=64]
tmpdir                /tmp
shell                 /bin/csh
prolog                NONE
epilog                NONE
shell_start_mode      posix_compliant
starter_method        /usr/local/bin/sge_cpuset
suspend_method        NONE
resume_method         NONE
terminate_method      NONE
notify                00:00:60
owner_list            NONE
user_lists            NONE
xuser_lists           NONE
subordinate_list      NONE
complex_values        num_proc=64
projects              NONE
xprojects             NONE
calendar              NONE
initial_state         default
...
 

-----Original Message-----
From: jgrn307 [mailto:greenberg at ucdavis.edu] 
Sent: Monday, January 11, 2010 1:07 PM
To: users at gridengine.sunsource.net
Subject: [GE users] Increasing slots past the number of CPUs?

Gridengineers:

    How can I increase the number of available slots to run jobs higher
than the number of CPUs we have on the system?  We've found in the past
that by increasing the # jobs to maybe 10% more than the # of CPUs we
get the fastest full-job execution, but I'm having a problem finding
where to set this parameter (in the "Parallel Environment Configuration"

I see the slots parameter is set to 999, but only 16 jobs (we have 16
cpus) execute at the same time, so clearly there's another parameter
hiding that needs to  be set).  Thanks!

--j

-- 

Jonathan A. Greenberg, PhD
Postdoctoral Scholar
Center for Spatial Technologies and Remote Sensing (CSTARS) University
of California, Davis One Shields Avenue The Barn, Room 250N Davis, CA
95616
Phone: 415-763-5476
AIM: jgrn307, MSN: jgrn307 at hotmail.com, Gchat: jgrn307

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessage
Id=238149

To unsubscribe from this discussion, e-mail:
[users-unsubscribe at gridengine.sunsource.net].

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=238152

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list