| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Retired Forums
Forum: Member-to-Member Support [Read Only]
Thread: Work unit seems too long |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
My agent has been working on the same piece of work for over 25 hours now and is still only about 50% complete. Is this normal? Thank you.
|
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well, it is only a little unusual. Depending on just how it folds, proteins can take 10 times as long as you would expect. Occasionally you will see this sudden change within a work unit, as it speeds along normally and then slows down dramatically when the folds switch to a different configuration.
mycrofth |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Back on about January 21, 2005 some very difficult work units were being handed out. I have a Pentium 4 3.2 Ghz 800 Mhz Front Side Bus with 1 GB memory and it took me six days to complete one work unit. Some work units can be completed in as short of time as three hours while others with difficult geometries, though fairly uncommon, can take days. 25 hours to complete 1/2 of a work unit is a bit long though not all that unusual. A friend of mine has a 1 Ghz machine and he just completed today a work unit that took 48 hours. The work unit that followed was completed in 6 hours.
|
||
|
|
Viktors
Former World Community Grid Tech Joined: Sep 20, 2004 Post Count: 653 Status: Offline Project Badges: 
|
Yes, these longer work units are normal. We are in a middle of a batch of genes which seem to be particularly difficult to fold, so the run times for some of these work units are running longer than usual. Unfortunately, we cannot accurately predict the difficulty ahead of time, so that is why there is variation in run time length. Our formula for tuning the work units did predict that these genes are more difficult, so the work was chopped into about 50% more work units than the usual average. But, it looks like they are even more difficult than our formula predicted. This is all ok, however. It will take a bit longer to get to sending the results, but you will get more points for the additional work at that time. Early in the project we had some genes that ran over 200 hours for one work unit.
|
||
|
|
|