| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Outsmart Ebola Together
Thread: Run time way to long |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 48
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I just aborted a OET work unit that was at 15% with over 7.5 hours CPU time behind it. Elapse time was like 7 hr 50 min....I know the new work units will take longer just because but it was set to take 4 hr 23 min on my CPU......okay so its not all that fast.
----------------------------------------Thing is the work buffer was already full with 6 days. The other work units seem okay with say 38% in 2.25 hours. That I can handle but with the time remaining having counted down to 42 min then up to 4 hours and not slowing down and just counting up..??..it got aborted. Didn't want to do it, felt like taking a old dog for a walk and coming back alone with one less shell in the shot gun. Had to be done. Is anyone else having that happen?? [Edit 2 times, last edit by Former Member at Feb 13, 2015 7:25:35 AM] |
||
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges: 
|
One of the first OET WUs I crunched this week lasted over 12 hrs. But then my queue is only one day. I would suggest reducing your queue to be able to include these big WUs.
----------------------------------------CJSL Crunching like there's no tomorrow... |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
some of mine take over 19 hours.....leave them to run!
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
In looking back, it's so good the project was -not- auto opted in. With the limited work availability [not had any new again since batch 345 at 02:02 UTC] all will get done, no feeder backlog of days or weeks.
19 hours on a PC, my tablet is now at 24 hours for a 333. So far none have completed in an unfavorable ending.  |
||
|
|
Barnsley_Tatts
Senior Cruncher Joined: Nov 3, 2005 Post Count: 281 Status: Offline Project Badges: 
|
The run times seem to be all over the place - I've had some that have taken >10 hours, some that have taken just an hour.
----------------------------------------Hopefully as the project settles down, the run times will standardise to something more manageable.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Nope, the bulk of OET work is single job 'flex' type, uncuttable/indivisible. Shortest here on one host 2:39 minutes, longest > 24 hours.
Even if the techs are able to figure out a pattern, which they said they were going to look for in the returned results, don't see them being able to split work up, other then by creating a sub group in the device profile to opt in for the big work. Modding the device profile for greater granular choice is only an act for [absent] higher powers, as most all know who loiter these forums.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
And a 333 on a 4770K just passed 26 hours and did a checkpoint 28 minutes ago, #6 :O
|
||
|
|
ErikaT
Former World Community Grid Admin USA Joined: Apr 27, 2009 Post Count: 912 Status: Offline Project Badges:
|
Hello wonderful members!
There have been some posts in this thread that have had to be removed. As a friendly reminder - please be kind and helpful to each other, and do not use profanity. And if you want to brush up on your forum etiquette, here are the forum rules. Many thanks for keeping it clean ,ErikaT |
||
|
|
jay_Orlando
Senior Cruncher USA Joined: Jan 4, 2006 Post Count: 181 Status: Offline Project Badges: 
|
Hi,
----------------------------------------I agree with CJSL and others. I just finished a wu that ran over 13 hours. I also suggest keeping a smaller net buffer. For example, I keep between 1.0 and 0.1. "Expectation Management" vs. Curiosity.... I would like to know more about Vina and the code for the project - just curious. I saw the short WU crunch times, and fell into the false expectation that all Ebola wu would be like that. I joined BOINC to have fun, learn more about programming, and to learn more about science. WCG fills those expectations for me. Jay --edit-- I read SekeRob's post about the feasibility of a queue for short jobs. That answered my question.  [Edit 1 times, last edit by jay_Orlando at Feb 13, 2015 4:58:26 PM] |
|||
|
||||
|
KWSN-A Shrubbery
Senior Cruncher Joined: Jan 8, 2006 Post Count: 476 Status: Offline Project Badges: 
|
I keep my queue at 1.25 days. Even with the variability nothing will stay so long as to not return in time. Wouldn't recommend this if you have systems that are older or hyperthread with average clock speeds.
----------------------------------------Granted, my projected run time can vary up to projecting 12-16 days but it simply stops requesting new work and eventually smaller work units bring that estimation back down. Guess it's possible to get a series of long running units and miss the deadline. I'll take the risk because living in the boonies sometimes connections are unavailable.  |
||
|
|
|