###############################################################################
#               Australian Partnership for Advanced Computing                 #
#                           National Facility                                 #
#      This service is for authorised clients only. It is a criminal          #
#      offence to:                                                            #
#                - Obtain access to data without permission                   #
#                - Damage, delete, alter or insert data without permission    #
#      Use of this system requires acceptance of the Conditions of Use        #
#      published at http://nf.apac.edu.au/policies/                           #
###############################################################################
|                 Welcome to the APAC National Facility!                      |
|    ac.apac.edu.au - a 1928 processor (1.6GHz) SGI Altix 3700Bx2 cluster     |
|   Assistance: help@nf.apac.edu.au     Information: http://nf.apac.edu.au    |
===============================================================================

  WARNING - excessive IO to /short affects all; use /jobfs when appropriate

  Notes:
     * There is currently a restriction that jobs of > 8cpus must be
       a multiple of 8cpus.

     * CPU requests for shared memory jobs (OpenMP, Gaussian, ...)
       must be of the form -lncpus=N:N to avoid the job being
       distributed over multiple nodes.

     * copyq jobs accessing MDSS (e.g. using the mdss command) should
       include the option -lother=mdss or risk failing when MDSS is down.

     * Some users may be unaware of the flexibility we have with queue
       limits.  It is possible to give per-user and per-project variations
       to limits when justified - just contact help@nf.apac.edu.au

 Aug 26: 20% of AC (ac25-ac30) crashed at about 7am. These nodes were
         already draining for a 10am downtime to repair an interconnect
         problem.  This problem was the cause of the crash. These nodes
         had to be rebooted again later after another crash. Apologies.

===============================================================================