Message boards :
Number crunching :
Won't get additional units
Message board moderation
Author | Message |
---|---|
Kurt Schmucker Send message Joined: 11 Jan 00 Posts: 72 Credit: 130,823,400 RAC: 207 |
One of my clients won't get any additional units -- just the ones actively running. The prefs on this client are set to get 9 days of extra units, it doesn't get any. Any suggstions? |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
There's a limit of 100 CPU tasks and 100 tasks per GPU per computer so the settings that you are playing with won't work past those other limits. Cheers. |
Tom M Send message Joined: 28 Nov 02 Posts: 5124 Credit: 276,046,078 RAC: 462 |
One of my clients won't get any additional units -- just the ones actively running. It will poll the Seti server more often if you set your "extra days" to something like 0.1 days. Otherwise it won't check in for 9 days :) Also, as far as I know there are no older generation or even "current" generation Intel iGPU's except maybe the IRIS ones that won't slow the over-all processing of that computer down when you run iGPU calculation tasks. You can move those machines into a "cpu only" work group (default, home, school, work) and get more production from them. Tom A proud member of the OFA (Old Farts Association). |
rob smith Send message Joined: 7 Mar 03 Posts: 22199 Credit: 416,307,556 RAC: 380 |
Nine "extra days"??? The extra days setting doens't work the way that it suggests :-( What it in effect does is means you will load your caches (the first figure), then it will attempt to store the "extra days", and only check back for more work after that time. There is a hard limit of 100 tasks for the CPU and 100 tasks for each GPU on a given computer, if you reach that limit in the "normal" cache then you will not get any extra, your computer won't ask for any for another "extra days". If you wont to keep your caches as full as possible then set a mid-range for the normal chach (say 5 days), and a SMALL value for the extra days (say 0.1 or less). Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
OMG. No, that isn't the way it works. (and watch out for part 2 of this post) BOINC maintains a 'low water mark' and a 'high water mark' for the amount of work stored on your computer, ready to run. The first figure in the preferences dialog is the low water mark: the second is the amount by which the high water mark exceeds the low water mark. Whenever the work stock reaches the low water mark, BOINC will request sufficient work to reach the high water mark - 9 days hence, in this case. At SETI, it won't get that much, because of the restriction on tasks in progress mentioned by others. Which means that it will work down to the low water mark in much less than nine days. And at that point it will request more work. When it 'needs' it, not by the calendar. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
One of my clients won't get any additional units -- just the ones actively running.You don't say which client. Your account lists 8 computers. At the time I checked, every single one of them was filled to the maximum amount permitted - 100 for the CPU, plus 100 extra for each GPU. You couldn't possibly have clicked the button 'Show active tasks' in the BOINC Manager on that machine, could you? Then you would see the tasks actively running, but not the others. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.