| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Africa Rainfall Project
Thread: progress is not saved |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 16
|
|
| Author |
|
|
Igelwurst
Cruncher Germany Joined: Jun 29, 2015 Post Count: 23 Status: Offline Project Badges: 
|
Hi,
----------------------------------------yesterday I started crunching with a Windows system in addition to my androids. The first WU of the African Rainfall project was processed for 5 hours. When I started the computer this morning, the calculation started again at 0. The WU of the Covid project, which was calculated in parallel, followed on from yesterday's 5h. Do you have a tip for me, what could be the reason? With the already large WUs of the Rainfall project, it is all the more regrettable if the intermediate results are not saved. Thanks very much! Igelwurst  |
|||
|
||||
|
geophi
Advanced Cruncher U.S. Joined: Sep 3, 2007 Post Count: 113 Status: Offline Project Badges: 
|
There are 8 checkpoints over the course of a model run, every 12.5%. Is it possible you didn't make it to 12.5% in 5 hours? Also, do you have BOINC configured to leave applications in memory when suspended? If not, that could be a reason why it didn't reach the first checkpoint.
|
||
|
|
Umlauf
Advanced Cruncher Joined: Mar 20, 2020 Post Count: 52 Status: Offline Project Badges: 
|
Hello Igelwurst,
ARP only has checkpoints every 12.5 %. Umlauf |
||
|
|
Igelwurst
Cruncher Germany Joined: Jun 29, 2015 Post Count: 23 Status: Offline Project Badges:
|
Hi geophi & Umlauf,
----------------------------------------thx for the quick answer. Probably I didn't reach the checkpoint. Good to know. Thanks! Igelwurst |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges: 
|
Igelwurst
ARP is not suitable if you are shutting down frequently. Every time you shut down it reverts to the previous 12.5%, which, in your case, was 0. Instead of shutting down, you could 'hibernate' or. 'sleep'. That would hold your work in memory. Mike |
||
|
|
Igelwurst
Cruncher Germany Joined: Jun 29, 2015 Post Count: 23 Status: Offline Project Badges:
|
Hi Mike,
----------------------------------------good hint - but I will probably use this machine to calculate OPN or MCM Thanks! Igelwurst |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
You could try my suggestion just to find out how long it takes for each checkpoint before giving up on it. It might be that you almost make it each time.
Mike |
||
|
|
eeqmc2_52
Cruncher Joined: Oct 29, 2006 Post Count: 4 Status: Offline Project Badges:
|
I'm having the same issue, even with "leave in memory" activated. When a task is paused by another BONIC project needing to run, everything reverts back to the last checkpoint.
|
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12594 Status: Offline Project Badges:
|
I don't know what else you are running but is it possible that you don't have enough memory available?
Mike |
||
|
|
dionisiodiazmanzano
Cruncher Joined: Jun 22, 2021 Post Count: 1 Status: Offline |
I was disappointed, when I saw that all the work I had done on my Africa Rainfall Project team had been lost, I thought it was a mistake, I'm sure they will fix it. Now I know that it is a normal operation and that they are not going to do anything.
I am going to ask you a favour, make the following modification: the period of time to save the work done should be shorter. |
||
|
|
|