[hnc_invincible] new total CPU time limit for longjobs
- From: Thomas Hartmann <thomas.hartmann@xxxxxxxx>
- To: "hnc_invincible@xxxxxxxxxxxxx" <hnc_invincible@xxxxxxxxxxxxx>
- Date: Wed, 09 Apr 2014 11:00:25 +0200
hello,
effective tomorrow, there will be a new limit for longjobs of 3 days.
why are we doing this? basically, jobs that run forever are a problem
because of two reasons:
1. fair division of resources is worse the longer the jobs are.
2. when we need to do maintenance on a cluster node, we would want to
avoid killing your jobs. so we would have to disable the node (so no
more jobs get started there) and then wait until the last job has
finished. if a job takes 10 days, we would have to wait 10 days and the
node would be usable by others.
if you have any questions or comments, you are welcome to share your
thoughts with us.
best,
thomas
--
Dr. Thomas Hartmann
CIMeC - Center for Mind/Brain Sciences
Università degli Studi di Trento
via delle Regole, 101
38060 Mattarello (TN)
ITALY
Tel: +39 0461 28 2779
Fax: +39 0461 28 3066
Email: thomas.hartmann@xxxxxxxx
Homepage: http://sites.google.com/site/obobcimec/
"I am a brain, Watson. The rest of me is a mere appendix. " (Arthur
Conan Doyle)
----
You receive this message because you are using the HNC Invincible Cluster at
the CIMeC in Mattarello.
This list is used to communicate important announcements like maintenance,
reboots, downtimes, problems etc.... This is NOT a discussion list! If you
have problems, please refer directly to the Admiral or Vice Admirals.
If you do not use the HNC Invincible anymore and do not want to receive any
news on it, you can unsubscribe by writing a mail with "unsubscribe" in the
subject line to: hnc_invincible-request@xxxxxxxxxxxxx.
Other related posts:
- » [hnc_invincible] new total CPU time limit for longjobs - Thomas Hartmann
