Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • RHEL6 64bit OS x86 nodes
  • 2.2GHz Sandy Bridge CPUs
  • 16 cores per node
  • 64GB RAM per node
  • QDR (40Gb) Infiniband for MPI comms
  • 10Gb ethernet for SLAC networking

 

Queuemin. # coresmax. # coresdefault runtimemax. runtime
bulletmpi851215 mins7 days
bulletmpi-large513204815 mins1 day

 


There are 2 mpi public queues for parallel computing at SLACMPI computing on the bullet cluster, bulletmpi and bulletmpi-large.  They They are available to anyone with a SLAC unix account, but we monitor these queues a bit more closely and you will need to . Please send email to unix-admin to request access to these queues.  Jobs submitted to these queues use some of the same batch hosts that are in use by the general farm.  

  • bulletmpi allows for jobs to request between 8 and 512 cores
  • bulletmpi-large allows for jobs between 513 and 2048 cores

Queuemin. # coresmax. # coresdefault runtimemax. runtime
bulletmpi851215 mins7 days
bulletmpi-large513204815 mins1 day

Single slot jobs are not allowed in these queues.  There There is also a limit on the total number of cores (slots) in use by the bulletmpi and bulletmpi-large queues.  You can see what that limit is You can check the current slot usage and the slot limits by running the command blimits which shows  that there are 3072 slots total available to those 2 queues and at the moment of the command they are all blimits command. The output below shows the combined slot total for bulletmpi and bulletmpi-large is limited to 3072 slots. All 3072 slots are in use:

renata@victoria $ blimits -w

INTERNAL RESOURCE LIMITS:

...