NFS@Home always restarting after resuming task
Message boards :
Questions/Problems/Bugs :
NFS@Home always restarting after resuming task
Message board moderation
Author | Message |
---|---|
Send message Joined: 12 Jun 13 Posts: 3 Credit: 1,280 RAC: 0 |
Hi, I run multiple BOINC projects and I have set a global time limit after which BOINC would automatically switch between the different WUs. Unfortunately, my current NFS@Home WU restarts itself each time after resuming. The WU is running here for at least 4 days, reaching 80% several times, but then the switch seems to clear everything, which is quite annoying. Is there anything I could do here? I am on Mac OS 10.6.8 with BOINC 7.0.65, the WU is G3p706_920700_0 (16e Lattice Sieve 1.10). Thanks, Ãdám |
Send message Joined: 26 Sep 09 Posts: 218 Credit: 22,841,893 RAC: 243 |
Lack of memory to run 16e Lattice Sieve 1.10. |
Send message Joined: 12 Jun 13 Posts: 3 Credit: 1,280 RAC: 0 |
That sounds weird. I remember having about 1G+ free memory when this happened, although it was quite a few days ago, so perhaps I'm wrong. Anyway, I already canceled that WU. |
Send message Joined: 26 Sep 09 Posts: 218 Credit: 22,841,893 RAC: 243 |
That sounds weird. I remember having about 1G+ free memory when this happened, although it was quite a few days ago, so perhaps I'm wrong. Anyway, I already canceled that WU. But running 4 instances on your CPU the applications will saturate your 4 GB. You need at least 1.3 GB per instance. For me it is not weird, it is called lack of memory on your i7 CPU M 620. |
Send message Joined: 12 Jun 13 Posts: 3 Credit: 1,280 RAC: 0 |
There must be a misunderstanding here. As far as Activity Monitor tells me, I always run a single BOINC project at the same time (it seems that to run multiple projects simultaneously, I would need to let BOINC use more than 50% of my CPU, which is a lot -- now it's limited to 20%, with a 40% limit on each core). How is it then possible that there were 4 instances on the CPU? |
Send message Joined: 13 Aug 13 Posts: 9 Credit: 2,500,034 RAC: 0 |
4 instances running with 20% each. There is another way to limit this, probably closer to what you expected : On multiprocessors, use at most 1 processors This will limit the number of concurrent BOINC tasks to 1 instead of the share that each of 4 tasks gets. |