| 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: 5
|
|
| Author |
|
|
RTS48
Veteran Cruncher Bolivia Joined: Aug 2, 2009 Post Count: 1353 Status: Offline Project Badges:
|
I have posted this thread on the BOINC Agent Support Forum but as it affects SCC I am repeating it here.
----------------------------------------I have 3 MAC computers with a total of 20 cores. Before the Cloud migration I had a stock of ~ 35 WUs per core (700 in progress WUs) and this was increased to ~ 70 WUs per core just prior to the Cloud migration so that users would not run out of WUs during the (predicted) 48 hours that communication with BOINC would be unavailable. Now that the Cloud migration is permitting communication it appears that the number of WUs per core has been reduced to ~ 35 again and the 'extra' WUs will be crunched until the backlog once again reaches this figure. So far so good. However, the 'extra' WUs that were sent are approximately twice as long as those that were being crunched prior to the Cloud migration with the result that I am now sitting of a whole raft of WUs that will not complete on time. I am sure that I am not the only person affected in this way. Will projects (especially SCC) allow some leeway in late reporting while everyones backlog is cleared to it's normal level?
Rod Peel
Santa Cruz Bolivia South America , ![]() |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
RTS48,
All tasks that were 'In Progress' were extended a report_deadline of 48 hours. There is leeway on all projects. By in progress, that means the boinc database scheduled the work unit to be assigned to a members machine. That means you could have been in the process of downloading the files, already started processing them, or started the upload when we shut down the grid for the migration. Thanks, -Uplinger |
||
|
|
mhammond
Advanced Cruncher USA Joined: Dec 22, 2011 Post Count: 130 Status: Offline Project Badges:
|
... I am now sitting of a whole raft of WUs that will not complete on time... Hey there Rod, If you're concerned that they won't complete, why not just abort some of them? Better than using cpu just to possibly get no credit. Are you concerned that they won't be replaced? Not seeing the issue. regards, mike ![]() |
||
|
|
RTS48
Veteran Cruncher Bolivia Joined: Aug 2, 2009 Post Count: 1353 Status: Offline Project Badges:
|
RTS48, All tasks that were 'In Progress' were extended a report_deadline of 48 hours. There is leeway on all projects. By in progress, that means the boinc database scheduled the work unit to be assigned to a members machine. That means you could have been in the process of downloading the files, already started processing them, or started the upload when we shut down the grid for the migration. Thanks, -Uplinger Thanks Uplinger. I've been doing a few calculations and I think that I can now complete what I have. Sunday 21st is the critical day. If I complete all of the WUs due on that date (350 x 3h / 20 cores) before it arrives I think my system will settle down. There is of course always the 'Abort' option if WUs will not complete on time.
Rod Peel
Santa Cruz Bolivia South America , ![]() |
||
|
|
RTS48
Veteran Cruncher Bolivia Joined: Aug 2, 2009 Post Count: 1353 Status: Offline Project Badges:
|
... I am now sitting of a whole raft of WUs that will not complete on time... Hey there Rod, If you're concerned that they won't complete, why not just abort some of them? Better than using cpu just to possibly get no credit. Are you concerned that they won't be replaced? Not seeing the issue. regards, mike Thanks Mike. I have aborted WUs before when the BOINC manager gets a bit behind but I do treat that as the last resort particularly as I am trying to get my 5 year Diamond badge forSCC before the WUs run owt.
Rod Peel
Santa Cruz Bolivia South America , ![]() |
||
|
|
|