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

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 254
Posts: 254   Pages: 26   [ Previous Page | 17 18 19 20 21 22 23 24 25 26 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 523621 times and has 253 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Computing for Sustainable Water Problems Thread

I'm not sure knreed was talking about credits-claimed:crunch-time equity, so I'll post this anyway.

cfsw_ 3687_ 03687440_ 1--   611  Valid   6/12/12 22:59:24   6/13/12 05:58:13   0.93 (3335 secs)   21.8 / 39.0
cfsw_ 3687_ 03687440_ 0-- 611 Valid 6/11/12 09:15:30 6/12/12 22:57:51 1.08 (3905 secs) 56.2 / 39.0



I recall seeing discussions on how the 'granted' split is done, before... I'm not questioning that;
what I'm not getting is how/why the slower machine's claimed credit of 56.2 is 258% higher than the faster machine's 21.8 when their crunching times differ by only ~15%.

How's that work, exactly?

Thanks. :-)
[Jun 13, 2012 7:50:39 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: Computing for Sustainable Water Problems Thread

On judgement call I've lowered the Dashboard indicator to Medium priority [if 250,000 results per day can be called medium]. In Runtime share that's what it's been doing over multiple days (could all be depending on a feeder share reshuffle, or to let the scientist get adjusted to the thought that things are running 4x faster and better pull out their experiment design stops to maintain the flow).

Making it up (from what we know) as I go

--//--
[Jun 20, 2012 9:53:45 AM]   Link   Report threatening or abusive post: please login first  Go to top 
alver
Senior Cruncher
Joined: Nov 30, 2007
Post Count: 245
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Computing for Sustainable Water Problems Thread

I'm having a problem understanding whether the points allocated for this project are now (roughly) in line with the other projects, per CPU hour, now. Can anyone shed any light on that?

Thanks.
----------------------------------------


(previously known as 'proxima' on SETI, UD, distributed folding, FaD, and Rosetta)
[Jun 20, 2012 12:17:19 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: Computing for Sustainable Water Problems Thread

Not even going to try to explain... knreed is too "looking at the dynamics" [Wont share the secrets... like how to play the stock market, but bears and bulls would be an excellent analogy ;>) ]

--//--
[Jun 20, 2012 12:27:18 PM]   Link   Report threatening or abusive post: please login first  Go to top 
alver
Senior Cruncher
Joined: Nov 30, 2007
Post Count: 245
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Computing for Sustainable Water Problems Thread

Grin - OK, thanks! When I was running this project exclusively, my 'points' recent averages fell away a fair bit, so I'll keep this project active, but with a mix of others, for a while.
----------------------------------------


(previously known as 'proxima' on SETI, UD, distributed folding, FaD, and Rosetta)
[Jun 20, 2012 1:09:39 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: Computing for Sustainable Water Problems Thread

Here's what I'm showing for points per day average on the last few tasks... (at least 2 years on each except for C4SW)

3769/day C4SW

3655/day GFAM

2708/day C4CW

2546/day DSfL


Results per day is 3 to 5 times higher on C4SW than any other WCG tasks I've run. Probably due to the 0.5 to 2 hour work unit run times.
[Jun 20, 2012 6:10:19 PM]   Link   Report threatening or abusive post: please login first  Go to top 
EZ123
Cruncher
USA
Joined: Nov 23, 2007
Post Count: 10
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Computing for Sustainable Water Problems Thread

New Problem: Stalled work units

Over the last week, among the thousand-plus work units that were successfully crunched, there were a few which seemed to stall out. The elapsed time kept increasing, but the % progress did not. I was able to restart these work units by first suspending, then resuming calculation [if I hadn't done this, I suspect that they would have remained stalled indefinitely].

The result log of one of these work units is shown below. It started at 08:20:01 and stalled at 09:18:18. After I suspended and resumed calculations at 18:17:55, it restarted from the last checkpoint and completed at 18:23:39.

Overall, I was granted 0.93 hours of CPU time credit.

This particular computer is an Intel i7 950 running @ 3.07 GHz under Windows 7 Service Pack 1. The CFSW application version number was 612.

Has anyone else encountered this problem?


Result Name: cfsw_ 5511_ 05511110_ 1--

<core_client_version>6.10.58</core_client_version>
<![CDATA[
<stderr_txt>
[08:20:01] INFO:Beginning simulation: 1990:240:1181469433
[08:22:49] INFO: Finished tick number 4
[08:24:08] INFO: Finished tick number 9
[08:25:23] INFO: Finished tick number 14
[08:26:46] INFO: Finished tick number 19
[08:27:57] INFO: Finished tick number 24
[08:29:18] INFO: Finished tick number 29
[08:30:35] INFO: Finished tick number 34
[08:31:49] INFO: Finished tick number 39
[08:33:09] INFO: Finished tick number 44
[08:34:18] INFO: Finished tick number 49
[08:35:40] INFO: Finished tick number 54
[08:36:54] INFO: Finished tick number 59
[08:38:15] INFO: Finished tick number 64
[08:39:35] INFO: Finished tick number 69
[08:40:47] INFO: Finished tick number 74
[08:42:07] INFO: Finished tick number 79
[08:43:21] INFO: Finished tick number 84
[08:44:48] INFO: Finished tick number 89
[08:46:04] INFO: Finished tick number 94
[08:47:19] INFO: Finished tick number 99
[08:48:38] INFO: Finished tick number 104
[08:49:48] INFO: Finished tick number 109
[08:51:10] INFO: Finished tick number 114
[08:52:25] INFO: Finished tick number 119
[08:53:44] INFO: Finished tick number 124
[08:55:05] INFO: Finished tick number 129
[08:56:18] INFO: Finished tick number 134
[08:57:37] INFO: Finished tick number 139
[08:58:52] INFO: Finished tick number 144
[09:00:15] INFO: Finished tick number 149
[09:01:30] INFO: Finished tick number 154
[09:02:47] INFO: Finished tick number 159
[09:04:07] INFO: Finished tick number 164
[09:05:17] INFO: Finished tick number 169
[09:06:41] INFO: Finished tick number 174
[09:07:57] INFO: Finished tick number 179
[09:09:15] INFO: Finished tick number 184
[09:10:34] INFO: Finished tick number 189
[09:11:48] INFO: Finished tick number 194
[09:13:09] INFO: Finished tick number 199
[09:14:21] INFO: Finished tick number 204
[09:15:41] INFO: Finished tick number 209
[09:16:59] INFO: Finished tick number 214
[09:18:18] INFO: Finished tick number 219
[18:17:55] DEBUG: Restarting from checkpoint.
[18:17:55]PctComplete = 0.929167
[18:17:55]ticks:currentTick:modules:currentModule:restart:seed240:223:6:0:0:19010

[18:19:54] INFO: Finished tick number 224
[18:21:04] INFO: Finished tick number 229
[18:22:23] INFO: Finished tick number 234
[18:23:39] INFO: Finished tick number 239
CPU TIME = 3343.756308
18:23:39 (2976): called boinc_finish

</stderr_txt>
]]>
[Jun 22, 2012 12:17:32 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: Computing for Sustainable Water Problems Thread


Has anyone else encountered this problem?


I had one such WU about five days ago. Had to exit and restart boinc to get it running again. It restarted from the last checkpoint and then completed successfully.
[Jun 22, 2012 9:29:40 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: Computing for Sustainable Water Problems Thread

Has anyone else encountered this problem?


If you run BOINCTasks and check it once a day or so, the stalled work units are pretty-apparent.

e.g.


It seems to happen more often on Intel processors than AMD for me... on the linux machines I just run
# /sbin/service boinc-client restart

On windows machines I use File->Exit in the manager with the 'stop running' box checked in the confirmation dialog, then restart it (which restarts the service/client, too, in windows)... suspend doesn't fix it for me, usually.
Maybe because I have 'suspend to memory' selected in Preferences?

edit1: undocumented
edit2: changed hosting location
edit3: corrected edit count
----------------------------------------
[Edit 3 times, last edit by Former Member at Aug 11, 2012 1:54:09 PM]
[Jun 22, 2012 2:22:40 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: Computing for Sustainable Water Problems Thread

knock on wood. i have not had this happen yet...
[Jun 22, 2012 5:50:59 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 254   Pages: 26   [ Previous Page | 17 18 19 20 21 22 23 24 25 26 | Next Page ]
[ Jump to Last Post ]
Post new Thread