Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 13
|
![]() |
Author |
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Any one noticed an increase in CPU time in the past day or so?
|
||
|
Deluxe_Cabinets_And_Granite
Veteran Cruncher Joined: Oct 27, 2008 Post Count: 939 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
my AMD quads which were doing a WU in 3.5 - 4 hrs ea. saw a bunch that took 6.5 to 8 ea. But it seems to be back to normal now...
----------------------------------------![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Any one noticed an increase in CPU time in the past day or so? [Pick your smiley], See http://bit.ly/WCGDSL, as on 'average' the DSFL run times [light blue curve] are very well behaved compared to last year, but that's only because we nearly quadrupled the number of results processed in a day [dark blue curve], so any variation gets very quickly smoothed out. Down to the individual device, nothing much has changed... always the same same with this science. |
||
|
Deluxe_Cabinets_And_Granite
Veteran Cruncher Joined: Oct 27, 2008 Post Count: 939 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Rob, sorry for OT, but congrats on your Blue Snurk globe!
----------------------------------------![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
[OT Cntd], thnx yes, a little premature, still needing about 28 days... a rounding I suppose, or something that stops border liners to drop back when the next research launches [OT Fin].
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Scribe I also have noted my normal 10-12 hrs boxes bouncing to 18-22 hrs and back, seems to be the last of the 90 series and no discernable pattern
|
||
|
joeperry39@gmail.com
Advanced Cruncher USA Joined: Nov 22, 2006 Post Count: 140 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I've noticed recently that jobs (most any project) that used to run in about 6 to 8 hours are now showing projected run times of 24 to 40 HOURS!!! And, the actual run times are in that general timeframe.
----------------------------------------And that's on my faster quad-core box. The slower dual-core is running easily double the times in the past. With these kind of times required to finish jobs, it is hardly worth tying up the computer to run wcg projects. I don't know what has caused this run-time problem, but it is really frustrating. ![]() "Everything in moderation, including moderation" -- Mark Twain |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Your comps must be running on 5Volts instead of 110 or 220. Here as a reminder, that it is absolutely -not- the case with any WCG science, the WCGART chart displaying the runtime average per science for the last 12 months.
|
||
|
dango
Senior Cruncher Joined: Jul 27, 2009 Post Count: 307 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I saw few targets (within 100 batch) which were computing with 6-8 hours instead of 3-4 like others. I had many (long) tasks form this affected target -> reject them (abort).
example: DSFL_ 00100-13_ 0000019_ 0268 8.04 / 8.09 DSFL_ 00100-13_ 0000019_ 0462 7.40 / 7.48 DSFL_ 00100-13_ 0000019_ 0172 8.19 / 8.25 and many others from _13_ but: DSFL_ 00100-10_ 0000035_ 0902 2.80 / 2.82 so yes I agree, that users can see longer times for some targets |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Individual machines with individual tasks that may run long, and -that- the techs can't do little to nothing about it. WCG plan for an average run time cutting and as the chart displays, that's well close to the objective.
To share again, sizes in 2 or 3 different classes are in the making [the server-code and work generation code is being developed now]... then the weaker brethren can also get to do tasks in that 6-8 range... with the unavoidable outliers [turning result challenges and rankings into an even bigger mockery]. Not sure if getting the right class size will require client v7, doubt it, but would be ready to upgrade when the word comes, via mail, via notice, via news and social connect channels. |
||
|
|
![]() |