Upd:Linux Cluster: UV still unavailable, MPP returned to operation

LRZ aktuell publish at lrz.de
Mi Feb 6 12:53:08 CET 2013

 Changes to this message: MPP returned to user operation
Update (Feb 6, 13:00): The MPP cluster has been returned to user
 operation. Please see below what you need to do as a consequence of the
 configuration changes. Dear users of the Linux Cluster systems at LRZ,
 unfortunately the Ultraviolet systems are presently unavailable for
 user operation, because a defective NUMAlink component needs
 We'll keep you updated about the status, and of course are working
 toward a speedy return to user operation. Apologies for the disruption
 of services.
 Recent configuration changes:
   * The default MPI environment on the MPP cluster was changed from
     Parastation MPI to Intel MPI. However, the mpi.parastation module
     will remain available for legacy use until the end of 2013. On the
     sgi ICE and UV systems, the sgi MPI (mpi.mpt) will remain default. 
     We strongly recommend that you recompile your applications with
     Intel MPI. If you insist on still running Parastation MPI binaries,
     you need to add the following lines to your SLURM job scripts:
     module unload mpi.intel
     module load mpi.parastation/5.0/intel
   * The 8-way Myrinet Cluster has been retired from parallel
     processing, and the nodes were added to the serial processing pool.
     This implies that the partition "myri_std" in the SLURM cell "myri"
     is now unavailable.
   * For the serial queues, SLURM fair share scheduling was introduced.
     For the parallel queues, a combination of fair share scheduling and
     favoring large jobs is now active. This is to prevent a single user
     from monopolizing cluster segments for long times if there are many
     jobs in the queue.
   * New storage systems have been introduced for the WORK (==PROJECT)
     and SCRATCH file systems. Please note that LRZ has only migrated
     WORK data to the new file system; data in SCRATCH have not been
     migrated. However the old SCRATCH file system will remain available
     as a separate mount in read-only mode on the login nodes until the
     end of March, 2013. Migration of data can then be done via commands
     cp -a <my_scratch_subdirectory> $SCRATCH
     The environment variable $SCRATCH_LEGACY will remain defined and
     point to the legacy scratch area until end of March, 2013.

 This information is also available on our web server

 Reinhold Bader

Mehr Informationen über die Mailingliste aktuell