| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: BOINC Points post in Member News |
| No member browsing this thread |
|
Thread Status: Locked Total posts in this thread: 277
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for the information as always knreed
There are some that fall through the cracks though: faah0891_ bdb543_ mx7hvp_ 04 Valid 11/11/2006 04:22:31 11/11/2006 15:10:18 5.33 43 / 43 faah0891_ bdb543_ mx7hvp_ 04 Valid 11/11/2006 04:17:59 11/11/2006 23:05:42 5.36 43 / 43 faah0891_ bdb543_ mx7hvp_ 04 Invalid 11/11/2006 04:13:36 11/12/2006 17:22:51 34.75 105 / 22 This is and always has been a typical claim for the system. the work unit is probably of average duration... My feeling is that the 43 points are in fact low outliers, especially given that around 65 seems to be a fair average point expectation for an average WU. Surely the difference between the time taken to complete could be seen as a mitigating factor here. Cheers. ozylynx  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
May I ask, how the scoring system will work when the BOINC version of HPF2 gets going again soon, when there is no quorum ? What method will be used to compare work done, if for instance a fast machine gets a very complex job and a slow machine gets a less complex job and the slow machine completes quicker than the fast one ?
I do understand why the quorum is not necessary to validate the science with HPF2 BTW. |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
Ozylynx,
I'm looking at that computer and I'm wondering what else do you have running on that machine. You are close to the memory limit on that machine. We state that we need 250MB to run FightAIDS@Home on the system requirements page and it looks like that computer barely meets it. Is there any chance that you are paging pretty hard on that machine? That would significantly slow down the computers run time. Also - it would benchmark faster then it would run FightAIDS@Home since the benchmark does not use nearly as much memory. thanks, Kevin |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
May I ask, how the scoring system will work when the BOINC version of HPF2 gets going again soon, when there is no quorum ? What method will be used to compare work done, if for instance a fast machine gets a very complex job and a slow machine gets a less complex job and the slow machine completes quicker than the fast one ? I do understand why the quorum is not necessary to validate the science with HPF2 BTW. We have not written the validation/credit granting code for HPF2 on BOINC yet. Until We do and make sure that what we are planning works then we will refrain from providing details lest those details change. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
These are all pending from the D930 @3600
You will see they are almost as low as Ozzy`s yet WU`s are done quicker than average ! Workunit Name Device Name Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit faah0894_ bdb837_ mx8hvp_ dry_ 00 psycho-240756 Pending Validation 11/14/2006 20:21:48 11/15/2006 14:39:53 4.54 45 / 0 faah0894_ bdb805_ mx9hvp_ 05 psycho-240756 Pending Validation 11/14/2006 19:12:12 11/15/2006 11:27:47 4.62 46 / 0 faah0894_ bdb637_ mx8hvp_ dry_ 00 psycho-240756 Pending Validation 11/14/2006 14:21:49 11/14/2006 23:14:12 4.63 46 / 0 B10275_ 0016_ CTMA1Aa-14-1-10 psycho-240756 Pending Validation 11/14/2006 10:22:08 11/14/2006 17:12:37 2.91 29 / 0 faah0894_ bdb094_ mx8hvp_ dry_ 03 psycho-240756 Pending Validation 11/13/2006 21:45:53 11/14/2006 17:12:37 4.57 45 / 0 faah0893_ bdb865_ mx7upj_ dry_ 04 psycho-240756 Pending Validation 11/13/2006 19:21:22 11/14/2006 14:21:49 4.83 48 / 0 faah0893_ bdb709_ mx8hvp_ 01 psycho-240756 Pending Validation 11/13/2006 14:33:02 11/13/2006 21:45:52 4.77 47 / 0 faah0893_ bdb185_ mx7upj_ dry_ 06 psycho-240756 Pending Validation 11/12/2006 21:38:46 11/13/2006 12:15:51 4.68 45 / 0 faah0892_ bdb317_ mx7hvp_ dry_ 02 psycho-240756 Pending Validation 11/11/2006 23:17:10 11/12/2006 14:51:23 4.89 47 / 0 faah0892_ bdb168_ mx7upj_ 02 psycho-240756 Pending Validation 11/11/2006 18:45:58 11/12/2006 10:22:53 4.74 45 / 0 faah0891_ bdb822_ mx7hvp_ 03 psycho-240756 Pending Validation 11/11/2006 12:33:18 11/11/2006 21:25:23 4.52 43 / 0 |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
knreed,
----------------------------------------would the Duration Correction Factor in the client_state.xml file not be an indicator, that the bench is overstated? cheers
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Ozylynx, I'm looking at that computer and I'm wondering what else do you have running on that machine. You are close to the memory limit on that machine. We state that we need 250MB to run FightAIDS@Home on the system requirements page and it looks like that computer barely meets it. Is there any chance that you are paging pretty hard on that machine? That would significantly slow down the computers run time. Also - it would benchmark faster then it would run FightAIDS@Home since the benchmark does not use nearly as much memory. thanks, Kevin Kevin: It has 256MB of ram. It only has 2 DIMM slots and 2x128 is all I have to put in it. If you are looking at your posted results well that's all there is. I bought RAM a while back to get it up to 256 I wont be buying more You might be looking at the WU it just submitted. It was running a different project waiting for changeover to that project. I'll change it back when this is sorted. On all previous work, such as the one I submitted, it is a headless drone working on WCG 24/7 exclusively. Cheers ozylynx |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Just a reminder to all.
We are trying to get 2 new projects out the door. Christmas is coming. HPF2 still needs a lot of work. About all we can do on points just now is gather statistics and think about what to recommend for a long term solution. It's a bit like steering an ocean-going freighter. You have to think a long way ahead. It is very different than steering a lightweight automobile. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, Lawrence it`s a fine balance looking after a workforce and the customers, one without the other doesn`t work at all ! ;-)
|
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
knreed, would the Duration Correction Factor in the client_state.xml file not be an indicator, that the bench is overstated? cheers The duration correction factor is going to vary based on whether you get longer running workunits or shorter running workunits which since they are randomly distributed (as we cannot predict the length in advance). Additionally, BOINC was originally written with the thought that 1 project (i.e. website/URL) = 1 science application. The duration correction factor doesn't vary by science application so when we are running 4 science applications at once the duration correction factor will have a hard time remaining accurate. |
||
|
|
|