checkpoint written ??

Message boards : Number crunching : checkpoint written ??

To post messages, you must log in.

AuthorMessage
_heinz

Send message
Joined: 30 Jun 06
Posts: 24
Credit: 38,697
RAC: 0
Message 20065 - Posted: 12 Jul 2006, 11:53:39 UTC
Last modified: 12 Jul 2006, 12:04:01 UTC

I stopped the client shutdown the pc and start again.
The Wu starts but not at the same point, when I stopped the client and shut down.
It looks like that the programm dont write a checkpoint if it is closed.
Know anybody something about it.
edit:
The wu was about 50% now it starts with 35% ---> a lot waste time
ID: 20065 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile KSMarksPsych
Avatar

Send message
Joined: 15 Oct 05
Posts: 199
Credit: 22,337
RAC: 0
Message 20069 - Posted: 12 Jul 2006, 13:06:17 UTC

I believe there are only certain points in the calculations that the Rosetta app is able to write the checkpoint.

I know one of those point for certain is at the end of a model. I'm not sure if or where it is able to checkpoint in addition to that.
Kathryn :o)
The BOINC FAQ Service
The Unofficial BOINC Wiki
The Trac System
More BOINC information than you can shake a stick of RAM at.
ID: 20069 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Feet1st
Avatar

Send message
Joined: 30 Dec 05
Posts: 1755
Credit: 4,690,520
RAC: 0
Message 20093 - Posted: 12 Jul 2006, 22:35:47 UTC

Yes Seti, that's the way it works. When you shutdown BOINC you will ALWAYS lose some work. The question is how much? And Rosetta has added additional checkpoints to try and minimize the work lost and bring it down to 0-20 min, for an average of 10 min. But with these large CASP proteins, they aren't able to checkpoint as often as every 20 min. so I've heard of people losing up to an hour of crunch time.

It's the nature of the beast. If I'm not mistaken, you do get credit for the lost time. But yes, it would be better for everyone if less time were lost. Unfortunately, during a competition is not the time to make such changes.

In your case, it looks like you are running the default 3hr work units, so you lost about 15 minutes of work when you went from 50% back to 35%. Compare the 15min to the total crunch time for the day and hopefully it proves to be a very minor percentage overall.

The ideas of shutting down AFTER the next checkpoint, and checkpointing more frequently have already been discussed elsewhere, so please don't feel you are being ignored. But do keep in mind that in the big scheme of things you are talking about a 1 or 2% deal here in most cases.
Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/
ID: 20093 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Number crunching : checkpoint written ??



©2024 University of Washington
https://www.bakerlab.org