Advanced search

Forums : Technical Support : New Work Units
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile [B^S] Gamma^Ray
Volunteer tester
Avatar

Send message
Joined: 7 Jun 07
Posts: 47
Credit: 70,587
RAC: 0
Message 191 - Posted: 13 Jun 2007, 22:05:10 UTC

Latest batch I ran worked fine, Out of wu's right now to test. :(
Windows-XP-Pro, AMD 3800X2, 5.10.28
ID: 191 · Report as offensive     Reply Quote
Profile [B^S] Acmefrog
Volunteer tester
Avatar

Send message
Joined: 8 Jun 07
Posts: 175
Credit: 446,074
RAC: 0
Message 192 - Posted: 13 Jun 2007, 23:09:52 UTC

They seem to be working now foe me as well. Also it looks like we ow need a quarum of 2 for each WU.
ID: 192 · Report as offensive     Reply Quote
Profile Scott
Volunteer moderator
Project administrator
Project developer
Avatar

Send message
Joined: 1 Apr 07
Posts: 662
Credit: 13,742
RAC: 0
Message 199 - Posted: 14 Jun 2007, 14:02:04 UTC - in response to Message 192.  

They seem to be working now foe me as well. Also it looks like we ow need a quarum of 2 for each WU.

You are correct. Up until now I've been using a 1-WU quorum and the sample bitwise validator. I just finished our *real* validator, so that's what I'm testing with these new WUs.
Scott Kruger
Project Administrator, Cosmology@Home
ID: 199 · Report as offensive     Reply Quote
Profile Rebirther
Volunteer tester
Avatar

Send message
Joined: 22 May 07
Posts: 23
Credit: 277,369
RAC: 0
Message 200 - Posted: 14 Jun 2007, 14:05:29 UTC

I think a quorum of 2 is not needed and wast time, fixed credits would be fine.
ID: 200 · Report as offensive     Reply Quote
John McLeod VII
Volunteer tester
Avatar

Send message
Joined: 8 Jun 07
Posts: 28
Credit: 305,469
RAC: 0
Message 205 - Posted: 14 Jun 2007, 20:07:24 UTC - in response to Message 200.  

I think a quorum of 2 is not needed and wast time, fixed credits would be fine.

That depends on whether the second task is required to double check to see that the work was done correctly. Not all tasks are amenable to a quick test for correctness after the answer is found.


BOINC WIKI
ID: 205 · Report as offensive     Reply Quote
Profile Scott
Volunteer moderator
Project administrator
Project developer
Avatar

Send message
Joined: 1 Apr 07
Posts: 662
Credit: 13,742
RAC: 0
Message 206 - Posted: 14 Jun 2007, 20:18:16 UTC

While I am told that CAMB is fairly numerically stable, we have to make sure that different architectures aren't sending back erroneous results. Therefore, we are double checking all of the test WUs.
Scott Kruger
Project Administrator, Cosmology@Home
ID: 206 · Report as offensive     Reply Quote
Profile Rebirther
Volunteer tester
Avatar

Send message
Joined: 22 May 07
Posts: 23
Credit: 277,369
RAC: 0
Message 207 - Posted: 14 Jun 2007, 20:31:04 UTC - in response to Message 206.  

While I am told that CAMB is fairly numerically stable, we have to make sure that different architectures aren't sending back erroneous results. Therefore, we are double checking all of the test WUs.


Ok, ok, scotty in the machine room ;)
ID: 207 · Report as offensive     Reply Quote
Profile Krunchin-Keith [USA]
Volunteer tester
Avatar

Send message
Joined: 21 May 07
Posts: 6
Credit: 132,291
RAC: 0
Message 208 - Posted: 14 Jun 2007, 22:31:35 UTC

I just discovered my linux host is having problems,

For Camb 1.12 it shows so far 6 results with an error 'Resource Limit Exceeded'

and when checking the result it shows something like:

<core_client_version>5.8.17</core_client_version>
<![CDATA[
<message>
Maximum CPU time exceeded
</message>
<stderr_txt>

</stderr_txt>
]]>

They all pretty much did this between 2:03:58 to 2:10:36 run time.

It does show success for 10 results with shorter times form 00:05:32 to 00:17:10
ID: 208 · Report as offensive     Reply Quote
Profile Jord
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 Jun 07
Posts: 345
Credit: 50,500
RAC: 0
Message 265 - Posted: 19 Jun 2007, 8:23:31 UTC - in response to Message 206.  

While I am told that CAMB is fairly numerically stable, we have to make sure that different architectures aren't sending back erroneous results. Therefore, we are double checking all of the test WUs.

Very well done. I then assume my result wasn't numerically stable? WUID 80161. The only one so far.
ID: 265 · Report as offensive     Reply Quote
Profile Scott
Volunteer moderator
Project administrator
Project developer
Avatar

Send message
Joined: 1 Apr 07
Posts: 662
Credit: 13,742
RAC: 0
Message 271 - Posted: 19 Jun 2007, 14:30:13 UTC - in response to Message 265.  

While I am told that CAMB is fairly numerically stable, we have to make sure that different architectures aren't sending back erroneous results. Therefore, we are double checking all of the test WUs.

Very well done. I then assume my result wasn't numerically stable? WUID 80161. The only one so far.

Numerical stability has to do with the application, not the result.

The problem is that different architectures perform floating-point math differently, which will cause applications to produce different results from the same input data. This makes it difficult to decide which results are valid and which are not.

We're currently working on decreasing the rate at which results are invalidated.
Scott Kruger
Project Administrator, Cosmology@Home
ID: 271 · Report as offensive     Reply Quote
Profile [AF>EDLS>Ouest]_Damien
Volunteer tester
Avatar

Send message
Joined: 22 May 07
Posts: 7
Credit: 81,829
RAC: 0
Message 433 - Posted: 26 Jun 2007, 17:02:05 UTC

60 new WU for my computer during 2 hours! Isn'it too much?
ID: 433 · Report as offensive     Reply Quote
John McLeod VII
Volunteer tester
Avatar

Send message
Joined: 8 Jun 07
Posts: 28
Credit: 305,469
RAC: 0
Message 444 - Posted: 26 Jun 2007, 21:49:58 UTC - in response to Message 433.  

60 new WU for my computer during 2 hours! Isn'it too much?

What version of BOINC?
What are your settings for connect every X and extra work?


BOINC WIKI
ID: 444 · Report as offensive     Reply Quote
Profile Jayargh
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 25 Jun 07
Posts: 508
Credit: 2,282,158
RAC: 0
Message 448 - Posted: 26 Jun 2007, 23:25:01 UTC

I just got a message Boinc 5.10.7 on a 5150 with 4 cpu's ...you have reached max of 50 wu a day per host-no new work from project. Hmmmmm
ID: 448 · Report as offensive     Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 25 Jun 07
Posts: 77
Credit: 5,001,188
RAC: 0
Message 452 - Posted: 27 Jun 2007, 1:32:26 UTC - in response to Message 448.  

I just got a message Boinc 5.10.7 on a 5150 with 4 cpu's ...you have reached max of 50 wu a day per host-no new work from project. Hmmmmm

Per host? Normally the limits are per thread.
Dublin, CA
Team SETI.USA
ID: 452 · Report as offensive     Reply Quote
Profile Scott
Volunteer moderator
Project administrator
Project developer
Avatar

Send message
Joined: 1 Apr 07
Posts: 662
Credit: 13,742
RAC: 0
Message 481 - Posted: 27 Jun 2007, 13:57:45 UTC

The current settings are:
500 total results per day
50 WUs in progress
20 WUs sent at a time
1 result per host per WU

I'm still trying to find the sweet spot for these numbers, so they will certainly change over time.
Scott Kruger
Project Administrator, Cosmology@Home
ID: 481 · Report as offensive     Reply Quote
Profile [AF>EDLS>Ouest]_Damien
Volunteer tester
Avatar

Send message
Joined: 22 May 07
Posts: 7
Credit: 81,829
RAC: 0
Message 490 - Posted: 27 Jun 2007, 17:31:56 UTC

All right! Thanks.

ID: 490 · Report as offensive     Reply Quote
Kenneth Larsen
Volunteer tester
Avatar

Send message
Joined: 21 Jun 07
Posts: 23
Credit: 286,821
RAC: 0
Message 492 - Posted: 27 Jun 2007, 18:30:16 UTC

I've been getting this message on all my Linux hosts this day:

Wed Jun 27 20:28:52 2007|Cosmology@Home|Message from server: No work sent
Wed Jun 27 20:28:52 2007|Cosmology@Home|Message from server: (there was work but it was committed to other platforms)


Are you only generating Windows work?
ID: 492 · Report as offensive     Reply Quote
Profile sysfried

Send message
Joined: 24 Jun 07
Posts: 114
Credit: 5,296,905
RAC: 0
Message 495 - Posted: 27 Jun 2007, 18:52:31 UTC - in response to Message 492.  

I've been getting this message on all my Linux hosts this day:

Wed Jun 27 20:28:52 2007|Cosmology@Home|Message from server: No work sent
Wed Jun 27 20:28:52 2007|Cosmology@Home|Message from server: (there was work but it was committed to other platforms)


Are you only generating Windows work?


na, I got this one on a windows host as well.... (posted a few moments in another thread)

sysfried
ID: 495 · Report as offensive     Reply Quote
Profile Scott
Volunteer moderator
Project administrator
Project developer
Avatar

Send message
Joined: 1 Apr 07
Posts: 662
Credit: 13,742
RAC: 0
Message 497 - Posted: 27 Jun 2007, 19:01:08 UTC

ID: 497 · Report as offensive     Reply Quote
Previous · 1 · 2

Forums : Technical Support : New Work Units