completion times

Message boards : Number crunching : completion times
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile S@NL marx0012
Volunteer tester

Send message
Joined: 20 Jan 00
Posts: 10
Credit: 424,271
RAC: 0
Netherlands
Message 6201 - Posted: 10 Jul 2004, 18:44:02 UTC

Normaly the set client finishes a WU in under 4 hours,boinc says 4hours and 25 minutes are needed.
So far so good....

But today I downloaded two workunits that need 26 hours and 25 minutes?????

Does anyone know whats with the increase in time needed?
ID: 6201 · Report as offensive
Pascal, K G
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 6210 - Posted: 10 Jul 2004, 19:01:55 UTC

They should process in a normal time. I have completed a few and that is what happens...



Seti@Home, it is really about the science, isn’t it!!
ID: 6210 · Report as offensive
Profile S@NL marx0012
Volunteer tester

Send message
Joined: 20 Jan 00
Posts: 10
Credit: 424,271
RAC: 0
Netherlands
Message 6220 - Posted: 10 Jul 2004, 19:54:49 UTC

Thank you.....

ID: 6220 · Report as offensive
Profile MAOJC

Send message
Joined: 31 Jan 00
Posts: 11
Credit: 991,339
RAC: 0
United States
Message 6231 - Posted: 10 Jul 2004, 20:34:21 UTC

yeah but it screws the cache up, no wonder the splitters caught up!
ID: 6231 · Report as offensive
Profile Sir Ulli
Volunteer tester
Avatar

Send message
Joined: 21 Oct 99
Posts: 2246
Credit: 6,136,250
RAC: 0
Germany
Message 6232 - Posted: 10 Jul 2004, 20:51:30 UTC - in response to Message 6231.  

> yeah but it screws the cache up, no wonder the splitters caught up!
>
>

yes and then low-water mark, and high-water mark is useless...or getting in disorder.


Greetings from Germany NRW
Ulli [/url]
ID: 6232 · Report as offensive
Profile adrianxw
Avatar

Send message
Joined: 14 Jul 99
Posts: 173
Credit: 1,698,756
RAC: 3
Denmark
Message 6235 - Posted: 10 Jul 2004, 21:03:22 UTC

>>> Seti@Home, it is really about the science, isn’t it!!

Believe that, and you've alienated vast numbers of crunchers. It is the competitive element that keeps people on board.
<p>
------------------
Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.


ID: 6235 · Report as offensive
Pascal, K G
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 6236 - Posted: 10 Jul 2004, 21:05:47 UTC
Last modified: 10 Jul 2004, 21:06:20 UTC

Well the system is working, as the last WUWUtm gets to about 50% completed, on my P4 1.8, BOINC downloads 1 more WUWUtm, with a completiom time of 41 hrs (they complete in a normal time frame).
On the p4 3.0 I am still waiting for my 5 WUWUstm to crunch, will report back what happens when I hit the low water mark......;o)



Seti@Home, it is really about the science, isn’t it!!
ID: 6236 · Report as offensive
Pascal, K G
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 6286 - Posted: 10 Jul 2004, 23:41:31 UTC

THe p4 3.0 is being kept at 5 WUWUs, so I will be crunching steady.




Seti@Home, it is really about the science, isn’t it!!
ID: 6286 · Report as offensive
SURVEYOR
Volunteer tester

Send message
Joined: 19 Oct 02
Posts: 375
Credit: 608,422
RAC: 0
United States
Message 6317 - Posted: 11 Jul 2004, 1:13:01 UTC
Last modified: 11 Jul 2004, 1:13:19 UTC

From Rom
"Okay, it seems one of the SAH dev's noticed a trend on a large number of
machines where they were off by a factor of 6. So he adjusted the splitter
estimates by increasing them by a factor of 6."
Should have been only for Alpha wu's

BOINC Alpha Tester
BOINC Beta Tester
ID: 6317 · Report as offensive
Kerwin
Avatar

Send message
Joined: 7 Dec 99
Posts: 18
Credit: 10,635,562
RAC: 0
United States
Message 6344 - Posted: 11 Jul 2004, 3:17:06 UTC

Well I just hit my low water mark and I've gotten WUs with completion times that are in line with what I used to get before this 'anomaly'. I'm working through 4 units with completion times of 17:17:29. They've always finished at around 2:14:40. The new WUs that I just got have a completion time of 2:52:50.

I think everything's good for now.
ID: 6344 · Report as offensive
EclipseHA

Send message
Joined: 28 Jul 99
Posts: 1018
Credit: 530,719
RAC: 0
United States
Message 6351 - Posted: 11 Jul 2004, 3:58:57 UTC - in response to Message 6317.  
Last modified: 11 Jul 2004, 4:12:17 UTC

By fharmon but....

> From Rom
> "Okay, it seems one of the SAH dev's noticed a trend on a large number of
> machines where they were off by a factor of 6. So he adjusted the splitter
> estimates by increasing them by a factor of 6."
> Should have been only for Alpha wu's
>
> BOINC Alpha Tester
> BOINC Beta Tester
>

Ok... Maybe the reason is that the benchmarks are very, very wrong!

My Intel Cent 2.2g box running redhat has CONSISTANTLY been asking for 3-6 times the work of my P4 2.4g box on win2k! (same profile. since the beta, and I was told not to worry! Well, it's wrong even today!)
Both machine complete a WU within an hour of each other).

-BOINC Beta Tester

-Professional Programmer for 25 years.

-One who userstands College kids (profs/directors/dictators included) make mistakes, but also thinks they should be adult enough admit it, as that's the way they learn.

Your quote from from ROM (I can't find the message!) to me says "attack the symptom and not the cause! The only way this software will be ready is if the "cause" is corrected!


ID: 6351 · Report as offensive
SURVEYOR
Volunteer tester

Send message
Joined: 19 Oct 02
Posts: 375
Credit: 608,422
RAC: 0
United States
Message 6371 - Posted: 11 Jul 2004, 5:03:11 UTC

The Dev team is working on the soulution to the estimated completion time problem.
BOINC Alpha Tester
BOINC Beta Tester
ID: 6371 · Report as offensive

Message boards : Number crunching : completion times


 
©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.