Hello,

  While trying to troubleshoot performance issues on our NFS server, it
became apparent that we have a peak of requests durings hours 0-3 every
day. This is currently hard to properly manage and results in slower
performance for everybody.

  Here's a breakdown of the number of cronjobs scheduled to start in the
new SGE cluster, per hour:

     473 *  # hourly
     213 0
     106 1
      82 2
      78 3
      54 9
      50 4
      48 7
      45 12
      42 5

  You can see that, apart from hourly jobs, we have most cronjobs
starting in hours 1-3.

  My advice is that, if your workload permits, *schedule your cronjobs
outside of that time window so you are not competing for resources with
everybody else*. The more spread throughout the day your cronjobs are,
the better.

  Also beware of step values like "*/2" which will make your cronjob run
on hours 0 and 2 too.

  If you have any questions, please find us on #wikimedia-cloud

Regards,

--
Giovanni Tirloni
Operations Engineer
Wikimedia Foundation

_______________________________________________
Wikimedia Cloud Services mailing list
Cloud@lists.wikimedia.org (formerly labs-l@lists.wikimedia.org)
https://lists.wikimedia.org/mailman/listinfo/cloud