climateprediction.net home page
New work Discussion

New work Discussion

Message boards : Number crunching : New work Discussion
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 46 · Next

AuthorMessage
Profile geophi
Volunteer moderator

Send message
Joined: 7 Aug 04
Posts: 1931
Credit: 41,487,636
RAC: 4,640
Message 55488 - Posted: 14 Jan 2017, 18:33:19 UTC - in response to Message 55487.  

Yes. The folders for successful and failed tasks are cleaned up in Linux. I've run a bunch of 2 month tasks for testing without any problems or errors.

I then ran a 2 year hadcm3s and there was an error with that, but not the folder clearing problem that previously plagued the Linux versions.
ID: 55488 · Report as offensive     Reply Quote
Profile JIM

Send message
Joined: 31 Dec 07
Posts: 1121
Credit: 20,460,788
RAC: 4,016
Message 55502 - Posted: 17 Jan 2017, 17:08:00 UTC - in response to Message 55488.  

I see that there were a small number of task released this morning. All gone now.
ID: 55502 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 2738
Credit: 3,388,112
RAC: 2,679
Message 55513 - Posted: 18 Jan 2017, 17:38:53 UTC - in response to Message 55502.  

And some more there now if you are quick!
ID: 55513 · Report as offensive     Reply Quote
BRG

Send message
Joined: 24 Dec 16
Posts: 15
Credit: 1,564,952
RAC: 0
Message 55514 - Posted: 18 Jan 2017, 18:24:54 UTC - in response to Message 55513.  

And some more there now if you are quick!


Not allowing me to get any work, I was reading somewhere that users are not to keep clicking "update" since stopped doing that and still no work :(
ID: 55514 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7233
Credit: 23,154,247
RAC: 341
Message 55515 - Posted: 18 Jan 2017, 18:38:53 UTC - in response to Message 55514.  

The exact wording is important.
Does it say "Project has no work"?

********

There is a back off timer intended to limit hoarding of tasks, so that it's fairer to all users.
The timer is nominally 1 hour, but is actually 1 hour and (I think), 30 seconds.
Any contact with the server either starts the timer, or resets it to 1 hour, hence the "Don't keep clicking the Update button" advice.

And this latest lot was only 2 small batches, so the 10 thousand plus computers lying in wait for work have quickly grabbed all of them.

The Server Status page only updates about every 3 hours.
ID: 55515 · Report as offensive     Reply Quote
BRG

Send message
Joined: 24 Dec 16
Posts: 15
Credit: 1,564,952
RAC: 0
Message 55516 - Posted: 18 Jan 2017, 18:48:18 UTC - in response to Message 55515.  

Ah, it shows on the server page that there is work. but if its 3 hour updates it will be long gone by now...

Don't think it does say project has no work, I click update after seeing 2000 plus tasks under WAH2 and it then went to the deferred for an hour thing-so at that point I left it alone!
ID: 55516 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7233
Credit: 23,154,247
RAC: 341
Message 55517 - Posted: 18 Jan 2017, 19:48:38 UTC - in response to Message 55516.  

The messages are in the Event log, which can be accessed in the manager's menu under Tools.
Or press Ctrl+Shift+E
ID: 55517 · Report as offensive     Reply Quote
BRG

Send message
Joined: 24 Dec 16
Posts: 15
Credit: 1,564,952
RAC: 0
Message 55518 - Posted: 18 Jan 2017, 20:17:35 UTC - in response to Message 55517.  

The messages are in the Event log, which can be accessed in the manager's menu under Tools.
Or press Ctrl+Shift+E


Couldn't make it up! I set other projects running with work thinking I have no chance of getting any work units, hey guess what! I get 6 units!


Just have to be patient I guess! Thanks for the help Les.

    18/01/2017 6:19:41 PM | climateprediction.net | update requested by user
    18/01/2017 6:19:44 PM | climateprediction.net | Sending scheduler request: Requested by user.
    18/01/2017 6:19:44 PM | climateprediction.net | Requesting new tasks for CPU
    18/01/2017 6:19:47 PM | climateprediction.net | Scheduler request completed: got 0 new tasks
    18/01/2017 6:19:47 PM | climateprediction.net | Project has no tasks available
    18/01/2017 7:20:25 PM | climateprediction.net | Sending scheduler request: To fetch work.
    18/01/2017 7:20:25 PM | climateprediction.net | Requesting new tasks for CPU
    18/01/2017 7:20:28 PM | climateprediction.net | Scheduler request completed: got 6 new tasks

ID: 55518 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 2738
Credit: 3,388,112
RAC: 2,679
Message 55521 - Posted: 19 Jan 2017, 8:18:52 UTC

so the 10 thousand plus computers lying in wait for work have quickly grabbed all of them.


And, I wonder how many cores each of those 10,000 computers has? My three active boxes have 8 between them but the two dual core machines are quite old. I would guess that the average might be above 4 now.
ID: 55521 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7233
Credit: 23,154,247
RAC: 341
Message 55522 - Posted: 19 Jan 2017, 8:44:36 UTC - in response to Message 55521.  

I came across one user recently who had over a dozen computers. A couple had only 2 cores, but of the rest, about half had 10 and the other half 20.
And a year or so back, I came across a computer with 64 cores.

And if each of these had a large cache ...
ID: 55522 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 2738
Credit: 3,388,112
RAC: 2,679
Message 55523 - Posted: 19 Jan 2017, 8:55:46 UTC - in response to Message 55522.  

I came across a computer with 64 cores.


Wasn't there a 64 core one on linux trashing everything it downloaded because of missing libs?
ID: 55523 · Report as offensive     Reply Quote
Profile JIM

Send message
Joined: 31 Dec 07
Posts: 1121
Credit: 20,460,788
RAC: 4,016
Message 55524 - Posted: 19 Jan 2017, 13:45:53 UTC - in response to Message 55521.  

so the 10 thousand plus computers lying in wait for work have quickly grabbed all of them.


And, I wonder how many cores each of those 10,000 computers has?


I’d say at least 4 cores. I’m running the project on 3 mid-range laptops with I-5 processors and each has 2 physical cores and 2 virtual cores due to hyper-threading. Each can run 4 tasks at once. That’s 12 cores running 12 tasks at once. It’s a long way from the days of single core machines running one task at a time.
ID: 55524 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 2738
Credit: 3,388,112
RAC: 2,679
Message 55525 - Posted: 19 Jan 2017, 21:42:49 UTC

And there were a few more today though by the time I saw on the server status that there were any they had all gone so to speak.
ID: 55525 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7233
Credit: 23,154,247
RAC: 341
Message 55526 - Posted: 19 Jan 2017, 23:07:20 UTC - in response to Message 55523.  


Wasn't there a 64 core one on Linux trashing everything it downloaded because of missing libs?


Could have been. There've been so many causing problems.
ID: 55526 · Report as offensive     Reply Quote
Profile SolarSurfer

Send message
Joined: 10 Dec 04
Posts: 15
Credit: 4,870,098
RAC: 0
Message 55528 - Posted: 20 Jan 2017, 15:55:06 UTC - in response to Message 54840.  

There's over 200,000 tasks out there somewhere, so some people must have stockpiles.



I have 30 in my "stockpile" that are just hanging out in the ether somewhere since they are no longer on the physical machine. Can they be reissued so some other lucky person can crunch them?
"Nothing will benefit human health and increase chances for survival of life on Earth as much as the evolution to a vegetarian diet."
- Einstein
ID: 55528 · Report as offensive     Reply Quote
Profile Iain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1019
Credit: 5,511,676
RAC: 4,392
Message 55529 - Posted: 20 Jan 2017, 16:49:01 UTC - in response to Message 55528.  

There's over 200,000 tasks out there somewhere, so some people must have stockpiles.



I have 30 in my "stockpile" that are just hanging out in the ether somewhere since they are no longer on the physical machine. Can they be reissued so some other lucky person can crunch them?

If you detach and reattach to the CPDN project in BOINC Manager then the models will be reissued unless they're the last attempt in that work unit. If the models are completed but haven't been acknowledged as that then there's nothing you can do - that particular problem appears to be some server-side issue.
ID: 55529 · Report as offensive     Reply Quote
bernard_ivo

Send message
Joined: 18 Jul 13
Posts: 380
Credit: 14,057,320
RAC: 2,676
Message 55530 - Posted: 20 Jan 2017, 16:52:02 UTC - in response to Message 55528.  


I have 30 in my "stockpile" that are just hanging out in the ether somewhere since they are no longer on the physical machine. Can they be reissued so some other lucky person can crunch them?


If the machine is still in use, try detaching it from CPDN project then re-attach and some of the ghosts may materialize for others to crunch.
ID: 55530 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7233
Credit: 23,154,247
RAC: 341
Message 55531 - Posted: 20 Jan 2017, 20:46:03 UTC - in response to Message 55528.  

Really old unfinished tasks are probably of no use now. That part of the research will most likely have finished, perhaps with the help of another small batch to cover the missing data areas.

But if you do detach, you can take the opportunity to also change to the new https address when re-attaching, which is https://www.cpdn.org.
And then spend forever changing all of your bookmarked urls as well. <Sigh> I didn't realise that I had so many.
ID: 55531 · Report as offensive     Reply Quote
BRG

Send message
Joined: 24 Dec 16
Posts: 15
Credit: 1,564,952
RAC: 0
Message 55533 - Posted: 21 Jan 2017, 0:35:39 UTC - in response to Message 55522.  

I came across one user recently who had over a dozen computers. A couple had only 2 cores, but of the rest, about half had 10 and the other half 20.
And a year or so back, I came across a computer with 64 cores.

And if each of these had a large cache ...


I have 24 cores but use 20 of them! Don't think I can bring myself to setup burp on the servers that would give me 192 cores and then same again as logical, 2 of those! Would like to know how grid gpus do with folding?!?!! WU's have taken me along time to fold however... so I imagine our servers would take longer...

Others discuss users holding onto work, is this because they have their system in burp set to store for example 20 days of work or more?
ID: 55533 · Report as offensive     Reply Quote
Profile SolarSurfer

Send message
Joined: 10 Dec 04
Posts: 15
Credit: 4,870,098
RAC: 0
Message 55534 - Posted: 21 Jan 2017, 4:14:22 UTC - in response to Message 55531.  

Really old unfinished tasks are probably of no use now. That part of the research will most likely have finished, perhaps with the help of another small batch to cover the missing data areas.

But if you do detach, you can take the opportunity to also change to the new https address when re-attaching, which is https://www.cpdn.org.
And then spend forever changing all of your bookmarked urls as well. <Sigh> I didn't realise that I had so many.


I just successfully detached and reattached with no new work requested as I'm concentrating on WCG right now. I did both through BAM, since that was how I had set it up originally, but BAM didn't use the new URL when I reattached so I'm not sure if Willy doesn't know about it or if he just didn't set it up properly like what happened with QCN's URL change. Should I make a BAM bug report about it?

I still see the tasks in my In Progress list.

The ghost tasks weren't due until July and August 2017 when I got them. I don't know if that means anything in terms of their shelf life.
"Nothing will benefit human health and increase chances for survival of life on Earth as much as the evolution to a vegetarian diet."
- Einstein
ID: 55534 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 46 · Next

Message boards : Number crunching : New work Discussion

©2020 climateprediction.net