Quick Workunit

Message boards : Number crunching : Quick Workunit
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Profile Jord
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 15184
Credit: 4,362,181
RAC: 3
Netherlands
Message 26119 - Posted: 13 Sep 2004, 22:47:45 UTC

SETI@home - 2004-09-14 00:08:19 - Starting result 26ap04aa.4387.6961.386086.165_0 using setiathome version 4.03
SETI@home - 2004-09-14 00:09:07 - Computation for result 26ap04aa.4387.6961.386086.165 finished

72 seconds. Must be a worldrecord. Too bad I can't check the unit on the site though, just wondering if I got a noisy unit again. But as it seems, they are still out there. ;)


----------------------
Jordâ„¢

<img><img><img>
ID: 26119 · Report as offensive
Spacebadger
Avatar

Send message
Joined: 25 Jul 04
Posts: 13
Credit: 11,306
RAC: 0
Canada
Message 26132 - Posted: 13 Sep 2004, 23:20:40 UTC

Yep i think youre right about noisy unit. I have had a couple that were done in the 20 second range.
ID: 26132 · Report as offensive
Profile Tim Rieck
Avatar

Send message
Joined: 3 Jun 99
Posts: 7
Credit: 398,993
RAC: 0
Canada
Message 26166 - Posted: 14 Sep 2004, 1:10:44 UTC

I have had a few short wu in the past and have no idea why.
But whats really weird is I just had two wu that took forever to crunch, one was over 30 hours and the one that followed was over 70 hours. They usually take 11-13 hours to crunch, and the one being worked on now looks normal again.
Can someone elaborate alittle for me about how/why this happens?
Thanks
ID: 26166 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 26330 - Posted: 14 Sep 2004, 11:12:24 UTC - in response to Message 26166.  

> I have had a few short wu in the past and have no idea why.
> But whats really weird is I just had two wu that took forever to crunch, one
> was over 30 hours and the one that followed was over 70 hours. They usually
> take 11-13 hours to crunch, and the one being worked on now looks normal
> again.
> Can someone elaborate alittle for me about how/why this happens?
> Thanks

Did you turn on the screen saver?

What was the angle range of the work units (There is a chart in the SETI FAQ that talks to this)

Did you join more projects?

When minimized, which tab was present when the BOINC Work Manager was minimizzed?

There may be more, but these are a few of the things that may be able to explain part of the differences... Click on my signature to get the documentation menu ...

<p>
Click Me!


ID: 26330 · Report as offensive
texasfit
Avatar

Send message
Joined: 11 May 03
Posts: 223
Credit: 500,626
RAC: 0
United States
Message 26335 - Posted: 14 Sep 2004, 11:22:13 UTC

>When minimized, which tab was present when the BOINC Work Manager was minimizzed?
>
>

Paul
Does this really make a difference? Which Tab should be present when minimized to the tray?
----------



Join the Overclockers.com SETI Team!
ID: 26335 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 26345 - Posted: 14 Sep 2004, 12:03:38 UTC - in response to Message 26335.  

> >When minimized, which tab was present when the BOINC Work Manager was
> minimizzed?
> >
> >
>
> Paul
> Does this really make a difference? Which Tab should be present when
> minimized to the tray?

I have no conclusive proof, just rumor control that if set to Disk or Work, the GUI keeps updateing the panes even while minimized. THis may be urban legend... I do not know for sure. I usually set mine to Projects now.

You can see some of the other factors in the SETI FAQ I wrote ... click on signature to get started ...
<p>
For BOINC Documentaion: Click Me!


ID: 26345 · Report as offensive
texasfit
Avatar

Send message
Joined: 11 May 03
Posts: 223
Credit: 500,626
RAC: 0
United States
Message 26347 - Posted: 14 Sep 2004, 12:10:56 UTC - in response to Message 26345.  

> > >When minimized, which tab was present when the BOINC Work Manager
> was
> > minimizzed?
> > >
> > >
> >
> > Paul
> > Does this really make a difference? Which Tab should be present when
> > minimized to the tray?
>
> I have no conclusive proof, just rumor control that if set to Disk or Work,
> the GUI keeps updateing the panes even while minimized. THis may be urban
> legend... I do not know for sure. I usually set mine to Projects now.
>
> You can see some of the other factors in the SETI FAQ I wrote ... click on
> signature to get started ...
> <p>
>
>

I usually have mine on the message tab when I minimize. Might just test this theory due to my curious nature.

I do refer to your documentation quit often. Have read almost all of the 7387 pages. :-) Very well done, by the way. Thanks!!
----------



Join the Overclockers.com SETI Team!
ID: 26347 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 26353 - Posted: 14 Sep 2004, 12:30:09 UTC - in response to Message 26347.  

> I do refer to your documentation quit often. Have read almost all of the 7387
> pages. :-) Very well done, by the way. Thanks!!

Keep reading ... it changes almost all the time.

Health permitting I do up to 12 hours a day working on it ... The big project underway right now is to finish most of the BOINC Web Site Owner's Manual updated with the available projects ...

If you want to help, criticize the work, it keeps me on my toes.

And it is only 7385 pages .... :)

The best news is that I am below 300 broken links (out of 14,000 and change) and almost half of the pages still pass W3C testing (I try to fix a couple a day to improve my numbers there too).

<p>
For BOINC Documentaion: Click Me!


ID: 26353 · Report as offensive
Profile PyroFox
Avatar

Send message
Joined: 5 Apr 03
Posts: 155
Credit: 213,891
RAC: 0
Canada
Message 26355 - Posted: 14 Sep 2004, 12:35:23 UTC

yup. had 4 of those recently. I thought it was due to the optimized code, but I guess not, must be a fresh batch of noisy units.

I'll try out the minimized tab thing too Paul, give me a week or so! (school...and processing time :P)

-Fox
[/url]
ID: 26355 · Report as offensive
Profile Tim Rieck
Avatar

Send message
Joined: 3 Jun 99
Posts: 7
Credit: 398,993
RAC: 0
Canada
Message 26393 - Posted: 14 Sep 2004, 15:20:55 UTC - in response to Message 26330.  

Thanks for your reply Paul,

> Did you turn on the screen saver?
No, I never use screensaver

> What was the angle range of the work units (There is a chart in the SETI FAQ
> that talks to this)
I have no idea about this angle range

> Did you join more projects?
No, I have no interest in the other projects

> When minimized, which tab was present when the BOINC Work Manager was
> minimizzed?
always on messages tab

> There may be more, but these are a few of the things that may be able
> to explain part of the differences...
Thanks again, but nothing has changed on this machine to make them crunch longer. I just thought they were odd that they required that much time compared to my average crunch. I guess it's really no difference, if you can get units that are really short, why not some that take longer.





ID: 26393 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 26394 - Posted: 14 Sep 2004, 15:21:12 UTC - in response to Message 26355.  

> yup. had 4 of those recently. I thought it was due to the optimized code, but
> I guess not, must be a fresh batch of noisy units.
>
> I'll try out the minimized tab thing too Paul, give me a week or so!
> (school...and processing time :P)

Ok, take your time ... I am sure I am still going to be here ...

The rumor is that the Work and the disk tab are the worst.

In theory, the panes should not be updated when minimized ...

Better news is that I am sort of caught up with one of the upgrades I was trying to get done (page counters and multi-project capabilities).

I still have LOTs of holes and missing images so there is plenty to do. This is the kind of thing that would go into the Performance FAQ if we can prove it out. Not that it is something that really bothers me at this time. I would much rather hear that the Mac/Linux GUI is done :)

Hint! Hint! Hint!

It does look like one of the other things I was interested in seeing is started too ... Good O'l Chuck ...

<p>
For BOINC Documentaion: Click Me!


ID: 26394 · Report as offensive
Profile Papa Zito
Avatar

Send message
Joined: 7 Feb 03
Posts: 257
Credit: 624,881
RAC: 0
United States
Message 26456 - Posted: 14 Sep 2004, 19:44:35 UTC - in response to Message 26166.  

> I have had a few short wu in the past and have no idea why.
> But whats really weird is I just had two wu that took forever to crunch, one
> was over 30 hours and the one that followed was over 70 hours. They usually
> take 11-13 hours to crunch, and the one being worked on now looks normal
> again.
> Can someone elaborate alittle for me about how/why this happens?
> Thanks
>
>

Another possibility that people seem to forget about...

BOINC is set to low priority in the list of things that are running on your system. This means that anything with a higher priority (which is almost everything) will pre-empt BOINC from running. So BOINC is idle while you're playing Doom or that virus you have is launching a DoS attack on the Pentagon. This is also true for any services you may have running in the background. If you add up all the little things that can take time away from BOINC, a work unit could conceivably get pushed up to that 30-70 hour range you're seeing.

Then again, I could just be talking out of my butt.




------------------------------------


The game High/Low is played by tossing two nuclear warheads into the air. The one whose bomb explodes higher wins. This game is usually played by people of low intelligence, hence the name High/Low.
ID: 26456 · Report as offensive
Profile PyroFox
Avatar

Send message
Joined: 5 Apr 03
Posts: 155
Credit: 213,891
RAC: 0
Canada
Message 26467 - Posted: 14 Sep 2004, 20:11:24 UTC - in response to Message 26394.  

> > yup. had 4 of those recently. I thought it was due to the optimized code,
> but
> > I guess not, must be a fresh batch of noisy units.
> >
> > I'll try out the minimized tab thing too Paul, give me a week or so!
> > (school...and processing time :P)
>
> Ok, take your time ... I am sure I am still going to be here ...
>
> The rumor is that the Work and the disk tab are the worst.
>
> In theory, the panes should not be updated when minimized

ok, quick question for 'ya paul, bu minimized do you mean that you can see BOINC in the taskbar? or the B in the system tray?

I can do both, but it'll take longer... and I still hate school :P

-Fox
[/url]
ID: 26467 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 26527 - Posted: 14 Sep 2004, 22:48:23 UTC - in response to Message 26467.  

> ok, quick question for 'ya paul, bu minimized do you mean that you can see
> BOINC in the taskbar? or the B in the system tray?

B in the taskbar only. Minimized it will refresh the panes because it is still "open"

> I can do both, but it'll take longer... and I still hate school :P

Be a teacher, then you will really hate it ... :)

Actually, I did teach at a couple of the local universities and it is the theing I still miss the most ...

Papa,

Nope, you have it right...

But I have 6 computers, with two, their only task in life is BOINC ... The Mac is my workstation, and I have two HT machines of which one is for playing games (the other USED to be also, but now only does BOINC and audio work).

The last one is just a file server with a 300 GByte IDE RAID for protecting files ...
<p>
For BOINC Documentaion: Click Me!


ID: 26527 · Report as offensive
TWolf

Send message
Joined: 24 Jun 04
Posts: 7
Credit: 24,966
RAC: 0
Netherlands
Message 26688 - Posted: 15 Sep 2004, 7:31:17 UTC

I think I had quite a lot of "quick WUs" last night.

Seti is running on my server, so I don't keep much track of it.
I do know that I had 50+ WUs on my server yesterday.
Today I see that I only have 4.

It is impossible that it did 46+ WUs in one night, since it normally takes 4 hours to crunch one.

Or does anybody else have an explanation for the 'missing' WUs?
AniWay, a Dutch magazine about anime & manga
ID: 26688 · Report as offensive
Profile Contact
Volunteer tester
Avatar

Send message
Joined: 16 Jan 00
Posts: 194
Credit: 2,249,004
RAC: 0
Canada
Message 27435 - Posted: 17 Sep 2004, 20:17:39 UTC

Bizarre. I was doing some informal testing for the 'minimized pane bug' mentioned in this thread, when i noticed that i had processed 4 quick WU's (under 10 minutes to complete). Only happened once or twice in my SETI@home Classic. First time occurence for any of my BOINC's. I believe these short WU's were the only ones that were started and completed while BOINC was minimized on Work Pane. Maybe a coincidence. But bizarre for sure.

--------------------------------------------------------------------------------2004-09-14 21:49:24 [---] Starting BOINC client version 4.08 for windows_intelx86
2004-09-14 21:49:24 [SETI@home] Project prefs: no separate prefs for home; using your defaults
2004-09-14 21:49:24 [Pirates@Home] Project prefs: no separate prefs for home; using your defaults
2004-09-14 21:49:25 [SETI@home] Host ID is 173222
2004-09-14 21:49:25 [Pirates@Home] Host ID is 977
2004-09-14 21:49:25 [---] General prefs: from SETI@home (last modified 2004-09-13 16:49:13)
2004-09-14 21:49:25 [---] General prefs: no separate prefs for home; using your defaults
2004-09-14 21:49:25 [SETI@home] Resuming computation for result 05my04aa.7961.20192.384654.175_2 using setiathome version 4.03
2004-09-14 21:49:48 [---] Suspending network activity - user request
2004-09-15 04:18:04 [SETI@home] Computation for result 05my04aa.7961.20192.384654.175 finished
2004-09-15 04:18:04 [SETI@home] Starting result 27au03aa.10127.208.172172.206_7 using setiathome version 4.03
2004-09-15 13:54:25 [SETI@home] Computation for result 27au03aa.10127.208.172172.206 finished
2004-09-15 13:54:25 [SETI@home] Starting result 08au03ab.28481.401.934640.47_7 using setiathome version 4.03
2004-09-15 18:48:01 [---] Resuming network activity
2004-09-15 18:48:01 [---] Insufficient work; requesting more
2004-09-15 18:48:01 [SETI@home] Requesting 51749 seconds of work
2004-09-15 18:48:01 [SETI@home] Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2004-09-15 18:48:04 [SETI@home] Started upload of 05my04aa.7961.20192.384654.175_2_0
2004-09-15 18:48:04 [SETI@home] Started upload of 27au03aa.10127.208.172172.206_7_0
2004-09-15 18:48:10 [SETI@home] Scheduler RPC to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
2004-09-15 18:48:10 [SETI@home] General preferences have been updated
2004-09-15 18:48:10 [---] General prefs: from SETI@home (last modified 2004-09-15 18:36:08)
2004-09-15 18:48:10 [---] General prefs: no separate prefs for home; using your defaults
2004-09-15 18:48:10 [SETI@home] Project prefs: no separate prefs for home; using your defaults
2004-09-15 18:48:10 [Pirates@Home] Sending request to scheduler: http://pirates.vassar.edu/cgi-bin/scheduler
2004-09-15 18:48:14 [SETI@home] Finished upload of 05my04aa.7961.20192.384654.175_2_0
2004-09-15 18:48:14 [SETI@home] Throughput 6616 bytes/sec
2004-09-15 18:48:14 [SETI@home] Started download of 06my04ab.10554.12081.192314.188
2004-09-15 18:48:25 [SETI@home] Finished upload of 27au03aa.10127.208.172172.206_7_0
2004-09-15 18:48:25 [SETI@home] Throughput 7677 bytes/sec
2004-09-15 18:48:25 [SETI@home] Started download of 04my04aa.14406.26641.592332.126
2004-09-15 18:50:21 [---] Insufficient work; requesting more
2004-09-15 18:50:21 [SETI@home] Requesting 286332 seconds of work
2004-09-15 18:50:21 [SETI@home] Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2004-09-15 18:50:39 [Pirates@Home] Sending request to scheduler: http://pirates.vassar.edu/cgi-bin/scheduler
2004-09-15 18:51:09 [SETI@home] Finished download of 06my04ab.10554.12081.192314.188
2004-09-15 18:51:09 [SETI@home] Throughput 2068 bytes/sec
2004-09-15 18:53:35 [SETI@home] Finished download of 04my04aa.14406.26641.592332.126
2004-09-15 18:53:35 [SETI@home] Throughput 1164 bytes/sec
2004-09-15 18:55:01 [---] Suspending network activity - user request
2004-09-15 23:27:39 [SETI@home] Computation for result 08au03ab.28481.401.934640.47 finished
2004-09-15 23:27:39 [SETI@home] Starting result 05my04aa.7961.20192.384654.199_2 using setiathome version 4.03
2004-09-16 09:08:17 [SETI@home] Computation for result 05my04aa.7961.20192.384654.199 finished
2004-09-16 09:08:17 [SETI@home] Starting result 05my04aa.7961.20192.384654.184_1 using setiathome version 4.03
2004-09-16 09:16:38 [SETI@home] Computation for result 05my04aa.7961.20192.384654.184 finished
2004-09-16 09:16:38 [SETI@home] Starting result 05my04aa.7961.20192.384654.177_1 using setiathome version 4.03
2004-09-16 09:26:06 [SETI@home] Computation for result 05my04aa.7961.20192.384654.177 finished
2004-09-16 09:26:06 [SETI@home] Starting result 05my04aa.7961.20192.384654.183_0 using setiathome version 4.03
2004-09-16 09:33:34 [---] Running CPU benchmarks
2004-09-16 09:33:34 [---] Suspending computation and network activity - running CPU benchmarks
2004-09-16 09:34:36 [---] Benchmark results:
2004-09-16 09:34:36 [---] Number of CPUs: 1
2004-09-16 09:34:36 [---] 797 double precision MIPS (Whetstone) per CPU
2004-09-16 09:34:36 [---] 1428 integer MIPS (Dhrystone) per CPU
2004-09-16 09:34:36 [---] Finished CPU benchmarks
2004-09-16 09:34:37 [---] Resuming computation and network activity
2004-09-16 18:59:55 [SETI@home] Computation for result 05my04aa.7961.20192.384654.183 finished
2004-09-16 18:59:55 [SETI@home] Starting result 05my04aa.7961.20192.384654.176_1 using setiathome version 4.03
2004-09-16 19:08:18 [SETI@home] Computation for result 05my04aa.7961.20192.384654.176 finished
2004-09-16 19:08:18 [SETI@home] Starting result 05my04aa.7961.20192.384654.185_0 using setiathome version 4.03
2004-09-16 19:16:39 [SETI@home] Computation for result 05my04aa.7961.20192.384654.185 finished
2004-09-16 19:16:39 [SETI@home] Starting result 09se03ab.10549.16274.629824.76_6 using setiathome version 4.03
--------------------------------------------------------------------------------


For more on 'minimized pane bug' see this possibly related thread:
a bug, feature??


ID: 27435 · Report as offensive
texasfit
Avatar

Send message
Joined: 11 May 03
Posts: 223
Credit: 500,626
RAC: 0
United States
Message 27493 - Posted: 17 Sep 2004, 23:08:07 UTC

I have noticed in the past couple of days that I have been having a couple of these wu's that finish in the sub 10 minute range. I first thought there may have been an error but when I checked the stderr.txt file there was nothing listed.

I started looking for these more closely and they are showing up on both of my Boinc PC's which leads me to believe that during all the software/hardware issues we have had there appear to be some bad wu's. We know that if there is static or interference that the program with stop processing the wu.

May have been the splitter or just the SNAP 18000 scheduler with some sort of corruption. We did have quite a few issues with MD5 errors over the past few weeks, also. Could all be related, but all this is just speculation on my part.
----------



Join the Overclockers.com SETI Team!
ID: 27493 · Report as offensive
texasfit
Avatar

Send message
Joined: 11 May 03
Posts: 223
Credit: 500,626
RAC: 0
United States
Message 27719 - Posted: 18 Sep 2004, 15:16:07 UTC

The quick wu's seem to have gone away. Anyone else noticed this??

A couple of things I noticed which may have caused these.
1.) The wu's that the under 10 min were from downloads of 9th/10th which was just after one of the hardware issues.
2.) I have upgraded to v4.09 from v4.05 and at that time started processing wu's that were downloaded on the 12th.
----------



Join the Overclockers.com SETI Team!
ID: 27719 · Report as offensive
Profile Papa Zito
Avatar

Send message
Joined: 7 Feb 03
Posts: 257
Credit: 624,881
RAC: 0
United States
Message 27766 - Posted: 18 Sep 2004, 19:47:12 UTC - in response to Message 27719.  

> The quick wu's seem to have gone away. Anyone else noticed this??
>
> A couple of things I noticed which may have caused these.
> 1.) The wu's that the under 10 min were from downloads of 9th/10th which was
> just after one of the hardware issues.
> 2.) I have upgraded to v4.09 from v4.05 and at that time started processing
> wu's that were downloaded on the 12th.
> ----------


Closest I've come were a couple WUs that were done in an hour and a half. Normally they take 3 hours.




------------------------------------


The game High/Low is played by tossing two nuclear warheads into the air. The one whose bomb explodes higher wins. This game is usually played by people of low intelligence, hence the name High/Low.
ID: 27766 · Report as offensive
Profile Julian

Send message
Joined: 19 Jul 99
Posts: 75
Credit: 2,470,428
RAC: 0
United Kingdom
Message 27800 - Posted: 18 Sep 2004, 21:43:04 UTC

I am not convinced this is a noisy unit problem I think it is something to do with a bug in Boinc or SETI. I have had the same thing for 4 days running on one machine out of the 15+ running at the moment. I have tried reinstalling and will look on Monday at deinstalling then deleting entire directory before reinstalling. I would say the odds of one system loading about 200 dodgy files is unlikely.

(Before I get it in the next for letting it grab and waste these units I teach and have had no free time since Tuesday and the system is some way on the site!

ID: 27800 · Report as offensive
1 · 2 · Next

Message boards : Number crunching : Quick Workunit


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.