Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 5
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 5414 times and has 4 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Long time between checkpoints instead of what is defined in preferences

Both my linux and windows clients of WCG make checkpoints of CEP tasks almost every 1h instead of 1m that is defined on the properties. Is this normal behavior? This causes huge waste of CPU time so I stopped crunching data for this project
[Jan 24, 2011 9:00:11 PM]   Link   Report threatening or abusive post: please login first  Go to top 
gb009761
Master Cruncher
Scotland
Joined: Apr 6, 2005
Post Count: 2982
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Long time between checkpoints instead of what is defined in preferences

Yes, unfortunately, for this particular project, checkpoints are only possible after every 'step' of the job - and thus, these CAN be very long - upwards of 2-3 hours on certain steps.

Hence, one of the reason's why it's an 'opt-in' project.
----------------------------------------

[Jan 24, 2011 9:18:00 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Long time between checkpoints instead of what is defined in preferences

Both my linux and windows clients of WCG make checkpoints of CEP tasks almost every 1h instead of 1m that is defined on the properties. Is this normal behavior? This causes huge waste of CPU time so I stopped crunching data for this project

Hi nmag,

As per gb009761, those checkpoints are only possible at the end of every "job" of which there are 16 in a CEP2 task. The "Write To Disk" setting is a limitation rather than an instruction i.e. the science checks at each checkpoint if it's allowed to write per client setting, if it is ready to do so. The intervals here are so large because the models are huge and of course you don't want it to save 1GB to disk every minute. Then you'll basically not able to use your computer so slow it will appear.

For lots more info on checkpoints and limiting 'waste', visit this FAQ:

--//--
[Jan 25, 2011 8:57:49 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Long time between checkpoints instead of what is defined in preferences

OK, thank you both.

Since I have 2 extra hds to store data and my OSs are on ssd, can I move the dir where BOINC stores data so I prevent flash wear?
[Jan 25, 2011 9:52:50 PM]   Link   Report threatening or abusive post: please login first  Go to top 
CandymanWCG
Senior Cruncher
Romania
Joined: Dec 20, 2010
Post Count: 421
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
smile Re: Long time between checkpoints instead of what is defined in preferences

Hi nmag,

sure you can. Just follow this thread and you'll get there: http://www.worldcommunitygrid.org/forums/wcg/...ead,30666_offset,0#309242

After all, it worked for me! wink

Cheers!
----------------------------------------
Knowledge is limited. Imagination encircles the world! - Albert Einstein



[Jan 29, 2011 10:19:34 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread