Message boards :
CMS Application :
CMS ignores zero resource share?
Message board moderation
Author | Message |
---|---|
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0 |
I am running BOINC 7.16.1 on an Ubuntu 18.04 machine (i7-8700 with 12 cores) and have set CMS as my only project on a given location with "zero resource share". However, over the course of three hours, I have now picked up nine CMS work units, even though I am still running WCG on all cores, and won't finish for a few hours. Is this a bug in LHC or BOINC? Or an undocumented feature? |
Send message Joined: 29 Aug 05 Posts: 998 Credit: 6,264,307 RAC: 71 |
I am running BOINC 7.16.1 on an Ubuntu 18.04 machine (i7-8700 with 12 cores) and have set CMS as my only project on a given location with "zero resource share". See for example this thread. Resource share is not an instantaneous setting, and it has its foibles. |
Send message Joined: 15 Nov 14 Posts: 602 Credit: 24,371,321 RAC: 0 |
See for example this thread. Resource share is not an instantaneous setting, and it has its foibles. OK, that explains it more or less. I have seen that happen in a limited way before - whenever I attach to Rosetta for the first time with zero resource share, it will send a single work unit immediately, but I had never seen it operate over several hours before. It will straighten itself out shortly it seems. You have a good memory. |
Send message Joined: 29 Aug 05 Posts: 998 Credit: 6,264,307 RAC: 71 |
|
©2024 CERN