| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Fight Childhood Cancer Project Forum
Thread: To Completion times getting longer and longer (target-9) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 14
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Ive been focusing on HFCC for a few weeks now. I've noticed that the "To completion" on the tasks seem to be getting longer and longer. They started at around 2-4 hours, then all of a sudden they are extending to 8-12 hours! No changes in computing power or obvious causes for slower computing power. I did add a few devices recently.
Is there some sort of intelligence built in to the scripts that send tasks to crunchers? I'm guessing that you have seen that tasks are coming back reliably, possibly too quickly so you are sending larger chunks? If my guess is correct do i also get credit for the equivalent results? Thanks!  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello NetApp_Captain,
Yes. I had not noticed until I read your comment, but the new HFCC work units that I am getting are at least 50% longer. There has been no announcement, so this may be a change in the script that chops up work into work units or a chance effect of working on a new range of data. My guess is that it was intentional for some reason or other. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
That darker red line in http://bit.ly/WCGART tells the story. The noon number is 6.27 hours, where 6 hours is the mean target time. As you can see in the chart the times vary.
The cut for HFCC has for a very long time been consistent: All dockings for 1 compound spread over 2 tasks.... the molecule library is the same as the previous 8 targets FAIK. --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sekerob, thanks for the additional chart info. Assuming that the source of your data is definitive and the calculations are based on the total sample size over 7 days.
Then I guess I assume that my devices are giving me different "to completion" times based on the automatic benchmarking and real competition for processing resources. I'll keep an eye on it. Thanks! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You may 'assume' anything you like. If the can does not contain what's printed on it, toss it.
--//-- |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7659 Status: Offline Project Badges: 
|
I have a older slower system which has done HFCC for quite some time. The tasks have quite a wide variability from 12 to 30 hours. I am thinking it just depends on the particular molecules being examined. Some are more complex than others, that's all.
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks Sgt.Joe!
|
||
|
|
rilian
Veteran Cruncher Ukraine - we rule! Joined: Jun 17, 2007 Post Count: 1453 Status: Offline Project Badges:
|
Yep - the run time may vary on she same machine
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi.
----------------------------------------Talk about long tasks, these two are the first of many i have received & yet to run they are the longest I've seen on this project so far, the last couple of weeks they have been around 4hrs on this rig. And i thought they were suppose to be getting shorter! HFCC_ target-9_ 00844577_ target-9_ 0000_ 0-- Black-AMD-X6 Valid 6/2/12 22:53:02 6/3/12 23:10:36 11.22 465.7 / 364.0 HFCC_ target-9_ 00846086_ target-9_ 0000_ 0-- Black-AMD-X6 Valid 6/2/12 22:53:02 6/3/12 23:10:36 11.87 492.8 / 384.5 [Edit 1 times, last edit by Former Member at Jun 6, 2012 2:24:23 AM] |
||
|
|
rewood1000
Cruncher USA Joined: Apr 25, 2011 Post Count: 48 Status: Offline Project Badges:
|
I am glad this subject got brought up as I was starting to worry about my machines going odd on me. My times were around 4-6 hours per unit until a few days ago and have now shot up to 12-15 hours.
I checked and all are being validated except for the usual PV percentage. I am going for the sapphire and have 20 more calendar days to attain it, so an anomaly like this got me worried. If it is system wide, I am less worried now. |
||
|
|
|