| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Conquer Cancer
Thread: Longer processing time |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 25
|
|
| Author |
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Think you may have had a 'new' HPF2 job before this with a very short initial run time estimate
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
 No I have only had 8 hours HPF2 units lately if I'm not mistaken. But you're right, looks like the estimated times are way off and the HCC-units are the normal 5 hours ones.  |
||
|
|
tvdsluis
Cruncher Joined: Apr 12, 2007 Post Count: 26 Status: Offline Project Badges: 
|
Well, my wu's still take way too long to finish.
In my case it's not only the estimate time, but the actual processing time as well. Every wu on every machine takes 25% longer to process. Strange that everyone else has wu's that process alot faster after the update. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Which project(?) Talking HPF2? As announced the amount of work packed into these jobs was increased by an average of 50% and deadline increased from 11 to 18 days. All my other projects have returned to normal lengths. That said, the times and batches can vary in duration depending on the toughness of the target being processed. See supplemental graphs post in the Start Here forum for periodic update on daily and sliding average runtimes or the graph in this thread a few posts up for current standings.
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Feb 27, 2008 9:00:54 AM] |
||
|
|
tvdsluis
Cruncher Joined: Apr 12, 2007 Post Count: 26 Status: Offline Project Badges:
|
I'm talking about hcc wu's, as this is the hcc forum thread.
Posts in this thread indicate that the performance of hcc wu's has greatly improved. So I still find it strange that I am the only one with an increase of 25% of hcc wu processing time. I have 3 different machines, so its not just one setup. Nothing has changes on these machines.(1 vista, 2 win2000) The wu's can't be changed either, because other people say that there processing performance has greatly impoved. I'm just trying to understand why i am the only one with a constant 25% increase in proceesing time on my hcc wu's. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
No you're not the only one as eluded earlier. Long as the quorum average is ok consider it within norm.
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Feb 27, 2008 10:03:05 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I too have longer HCC times on a quad by 25%. Does this mean the reward is also 25% higher. Seems only fair.
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Yes Sir, since all our now equally consistent processing these units, the quorum average award should automatically adjust per job, but really the hourly credit is not changing 1 iota. Let me pull a sample of the latest HCC job done on my formerly very slow P4HT... stand by.
----------------------------------------okay, the machines have nicknames, all jobs with science version 5.20: X0000041980149200411181350_ 1-- Beetle Valid 02/22/2008 21:20:54 02/27/2008 14:29:24 10.22 79.6 / 88.7 X0000041980203200411181350_ 0-- Beetle Valid 02/22/2008 21:20:54 02/25/2008 20:48:29 11.72 91.1 / 79.4 X0000041890564200411121713_ 1-- Beamer Valid 02/22/2008 20:30:47 02/25/2008 11:37:19 7.05 73.9 / 68.4 X0000041980255200411181349_ 0-- Beetle Valid 02/22/2008 21:20:54 02/24/2008 15:45:44 11.63 88.9 / 82.9 X0000041980273200411181349_ 0-- Beetle Valid 02/22/2008 21:20:54 02/24/2008 03:22:04 11.29 86.3 / 93.1 X0000041890466200411121715_ 1-- Beamer Valid 02/22/2008 20:29:53 02/23/2008 20:29:19 6.91 72.1 / 70.7 X0000041530243200411011110_ 2-- Ferrari Valid 02/22/2008 13:43:49 02/23/2008 06:21:07 4.72 71.5 / 73.1 X0000042421302200412110009_ 0-- Ferrari Valid 02/21/2008 10:11:52 02/21/2008 18:01:21 4.44 67.2 / 75.6 On Average been getting more credit than asked for: 630.6 / 631.9, but only a little 
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Feb 27, 2008 8:11:56 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
okay, the machines have nicknames Mine are named after my pets, past and present. Thanks for the reply, I kinda figured the credits would go up based on preset formulas. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello Buster Gunn,
BOINC claims points according to a simple formula: C(benchmark speed) * CPU time. So if the CPU time goes up, the claimed credit goes up. All the complexity is in the algorithm that tries to reconcile different claims for the same work unit and decide what to award everyone who completed that unit successfully. Lawrence |
||
|
|
|