| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Downloading multiple units |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 20
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Still sunny there Janet?
Got dark in Devon ages ago   |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello folks.  Another newbie refugee from Grid.org (RIP) here! Using the UD Monitor program I was able to store spare work units, which I could crunch while offline. Is there any equivalent here on the BOINC agent? I will be unable to go online for a few days next week, and I would like to line up a pile of work for the computer to chew over in the interim. Thanks in advance.............. Snap on the being a newbie grid.org refugee I used UD monitor too for the same reason. However, I am not using BOINC, but the UD agent. For me and others in a similar position, will UD monitor as we know and love it work here at WCG? (I ask, because the discussions on it I tracked down date back several years) thanks. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Splendid! The next unit is due to finish in an hour or so. There is a big 6 hour unit waiting, which should keep it busy until tomorrow morning. Thanks for you help, and goodnight from sunny England. G'night all Splendid indeed to see the upcoming job with a duration estimate ready to go and as you process more, BOINC learns and adapts. The default is 0.3 days (7.2 hours) work buffer for new members, which is enough to scale the longest maintenance runs (usually Sunday), or server problems and keep that cruncher busy.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You start it up by using the wcg batch file.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You start it up by using the wcg batch file. Thanks. Sorry, I realised I missed the earlier post which said that.  |
||
|
|
olympic
Senior Cruncher Joined: Jun 12, 2005 Post Count: 156 Status: Offline |
A couple things to keep in mind:
-----------------------------------------the WCG server will only send out 10 work units at a time to a single machine. So if you request several days worth of work, it may take a few communication attempts to get your queue filled up. Allow BOINC to fill the queue on it's own schedule because hitting the update button will only get you the "communication deferred" message. Also, BOINC can be finnicky, I've had a machine take nearly a week to fill up it's queue to the 3 days worth that I requested. -individual work units have deadlines for when they have to be returned to WCG ranging from 7-9 days. So you don't want to queue up more work than your computer can crunch before the deadline. As stated, the default of 0.3 days is enough to cover normal server outages and 3-3.5 days is the max the system seems comfortable with.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
A couple things to keep in mind: -the WCG server will only send out 10 work units at a time to a single machine. So if you request several days worth of work, it may take a few communication attempts to get your queue filled up. Allow BOINC to fill the queue on it's own schedule because hitting the update button will only get you the "communication deferred" message. Also, BOINC can be finnicky, I've had a machine take nearly a week to fill up it's queue to the 3 days worth that I requested. -individual work units have deadlines for when they have to be returned to WCG ranging from 7-9 days. So you don't want to queue up more work than your computer can crunch before the deadline. As stated, the default of 0.3 days is enough to cover normal server outages and 3-3.5 days is the max the system seems comfortable with. Its all true! The computer finished the first three units, and just after they were completed, in came 10 units. These add up to around 53 hours, so I don't quite have my full quota, but lets see what happens in the next 1 and 1/2 hours, when the next unit is completed. Thanks again Janet |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
The present job due days standard is 9 for all WCG BOINC projects. Beta's and rush jobs have shorter periods ranging from part of a day to 2-4 days. The very short due date jobs usually only go to machines that have an established 'reliability' record. The buffer of work adapts to the actual crunching. Pause the agent or close the PC regularly for 8 hours daily, and it will be factored into the so called Duration Correction Factor (DCF).
----------------------------------------The WCG(UD) agent has a hard CPU time limit of 14 days or 3 weeks Wall clock after which they will time out. The time the agent is not loaded does not count.... the jobs will though time out regardless after 2 calendar months at zero or ninety nine percent sans rancune. Edit: corrected days to read hours!
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 3 times, last edit by Sekerob at May 2, 2007 10:45:25 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
-individual work units have deadlines for when they have to be returned to WCG ranging from 7-9 days. So you don't want to queue up more work than your computer can crunch before the deadline. As stated, the default of 0.3 days is enough to cover normal server outages and 3-3.5 days is the max the system seems comfortable with. 2.82 is the max queue setting for a 7 day deadline, 3.76 is the max for a 9 day deadline. Work Buffer in the wiki. |
||
|
|
E. Frijters
Senior Cruncher The Netherlands Joined: Apr 26, 2007 Post Count: 228 Status: Offline Project Badges: 
|
I increased the profile so I should get multiple WU's to crunch (I want no internet connections during the night, but I want Boinc to continue working then)
----------------------------------------When I press "Update", the following happens: 02/05/2007 07:59:14|WCG|Reason: requested by project 02/05/2007 08:30:56|WCG|Sending scheduler request: Requested by user 02/05/2007 08:30:56|WCG|(not requesting new work or reporting completed tasks) 02/05/2007 08:31:01|WCG|Scheduler RPC succeeded [server version 509] 02/05/2007 08:31:01|WCG|Deferring communication for 5 min 3 sec 02/05/2007 08:31:01|WCG|Reason: requested by project And after 5 minutes, nothing happens... My question: Why is it stating that I'm not requesting new work? [Answer:] Propably because I resetted (0,0) the "Connect to network about every" parameter in my profile.... duhhh!
Former grid.org slave
----------------------------------------  [Edit 1 times, last edit by E. Frijters at May 2, 2007 9:50:29 AM] |
||
|
|
|