Search Mailing List Archives
[farmshare-discuss] Barley sharing algorithm?
vkl at stanford.edu
Sun Mar 17 14:40:25 PDT 2013
I wonder if there are any updates on the Barley sharing situation. I
recently found a time when the entire cluster was unused and submitted a
job for 256 nodes, with a projected completion time of 24 hours.
However, it seems that the queuing system will happily continue running
jobs even when nodes are oversubscribed, and my projected runtime then
went well beyond the 48 hr limit (my job appears to have been killed
now, which fortunately for this specific simulation doesn't matter). In
every cluster I've used, users have to specify an allotted runtime in
the queue submission script, and the nodes are completely dedicated to
the job for the duration of that runtime, or until the job finishes. Is
there such a policy for Barley?
More information about the farmshare-discuss