Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 14
Posts: 14   Pages: 2   [ Previous Page | 1 2 ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 2465 times and has 13 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Very long CPU time WUs

very buggy with variable runtime wu aka HCMD. Does confirm my thoughts that client side was being ignored for some "server side" logic. Thank you for the 80 wu per day limit else my SINGLE core CPU would have another 4-5 days to dig out of this hole, 1.5 day cache, normally 10-20 wu pulled 86 more before limit shutdown, next wu finished pushed DCF to 9 days, have not pulled a wu on this box for 4 days and have estimated 2 more days to go

Edit: appears that service side adjusting is only done on work request, not on reporting, as I am now over cached, I guess I will not know new dcf (per server) until client sees need to request work
----------------------------------------
[Edit 1 times, last edit by Former Member at Mar 24, 2012 8:02:50 AM]
[Mar 24, 2012 7:55:41 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Ingleside
Veteran Cruncher
Norway
Joined: Nov 19, 2005
Post Count: 974
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Very long CPU time WUs

Edit: appears that service side adjusting is only done on work request, not on reporting, as I am now over cached, I guess I will not know new dcf (per server) until client sees need to request work

Yes, the per-application, per-host estimates in v7 is based on your validated results for application/host-combination, and starts taking effect after 10 validated results for application/host-combination. These only affects the estimates on new work, not work already downloaded on client.

Having DCF on top of the server-estimates just means it takes much longer for things to stabilize, for 1st. the DCF is corrected, for so the server-estimates gets corrected, for so the DCF must re-adjust with the new server-estimates.
----------------------------------------


"I make so many mistakes. But then just think of all the mistakes I don't make, although I might."
[Mar 24, 2012 1:51:09 PM]   Link   Report threatening or abusive post: please login first  Go to top 
retsof
Former Community Advisor
USA
Joined: Jul 31, 2005
Post Count: 6824
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Very long CPU time WUs

Hi Raul,

Unfortunately BOINC sees all WCG sciences as 1 pot, meaning if any task [Beta inclusive] runs for instance twice as long from estimate, BOINC assumes all other tasks to also run twice longer than estimate [your SN2S jobs in this case]. It will normalize again... we all need to allow for BOINC to give it time.

That said, the high variability of WCG tasks, single sciences and across the board, does force to keep the task buffer low as else panic will break out. A 5 days cache suddenly can become 10 days cache by the estimations and then there's deadline trouble. Keep it no higher than 2 days and the issue will rarely crop up.

cheers.

--//--
If you have more than one computer, you can use the BOINC home, work, school, and default device profiles to make separations of WCG projects, with different queues on each.

Be careful. Once you have made project changes in the detail device profiles, don't make any project change to the overall project page. It would OVERRIDE this and put the same projects everywhere.
----------------------------------------
SUPPORT ADVISOR
Work+GPU i7 8700 12threads
School i7 4770 8threads
Default+GPU Ryzen 7 3700X 16threads
Ryzen 7 3800X 16 threads
Ryzen 9 3900X 24threads
Home i7 3540M 4threads50%
----------------------------------------
[Edit 1 times, last edit by retsof at Mar 26, 2012 1:36:10 AM]
[Mar 26, 2012 1:33:13 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Very long CPU time WUs

Hello retsof,
If you have more than one computer, you can use the BOINC home, work, school, and default device profiles to make separations of WCG projects, with different queues on each. Be careful. Once you have made project changes in the detail device profiles, don't make any project change to the overall project page. It would OVERRIDE this and put the same projects everywhere.
-- snip from retsof [Mar 26, 2012 1:33:13 AM] post, italicization-emphasis mine.
Yes indeed, device-profile is one mechanism to separate or otherwise abstract a device for interfacing with other mechanisms/entities (projects, and hopefully later, per science-applications: GPU-based, CPU-based, beta?). As for the overall project page and to make the whole idea of separation/abstraction work, yes, crunchers have to ensure that they select/attach a specific profile-Name to a specific device-Name (and next clicked on "Save"), else the settings indicated in the overall project page (as Saved) would apply to the specific device-Name instead.
;
[Mar 26, 2012 10:20:58 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 14   Pages: 2   [ Previous Page | 1 2 ]
[ Jump to Last Post ]
Post new Thread