| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project Forum
Thread: Points calculation |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could someone please explain how the points were calculated for this WU.
Result Name Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit E000003_ 328A_ 00002700a_ 1-- Valid 06/12/08 20:18:07 07/12/08 22:05:51 8.91 77.0 / 77.0 E000003_ 328A_ 00002700a_ 0-- Valid 06/12/08 20:15:20 08/12/08 00:00:19 14.19 197.0 / 77.0 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Type A work units require redundancy, so the credit is calculated in the same way as for Help Conquer Cancer.
The algorithm uses the recent average credit history of both computers to determine what to grant. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Type A work units require redundancy, so the credit is calculated in the same way as for Help Conquer Cancer. The algorithm uses the recent average credit history of both computers to determine what to grant. While it is possible that the 2 computers have the same RAC, it looks like it took the lower of the 2 not the average. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Our FAQ's must be extremely hard to find even with an index:
----------------------------------------http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=6105#120888 Net: The higher was considered an outlier.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
As a result, we are going to change how the 2nd part of the process works. Instead of selecting the credit that is closest to its history, we will average the recent average history's for the two computers. It just didn't look like 2 computers could have the same RAC and be working on the same WU and be more than 30% different in crunch time. But I guess it happened. Thanks. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I said recent average credit is used in the algorithm, but I gave no details of the algorithm itself.
It is not used in the traditional BOINC way. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I said recent average credit is used in the algorithm, but I gave no details of the algorithm itself. But the FAQ says the RAC (I make no assumption on the actual definition) for the 2 computers is averaged. Since the result is equal to the RAC for the faster computer, then the only way to get the posted result is for the RAC on both computers to be the same. Either the FAQ and the posted result are correct or both are incorrect and the result is calculated some other way. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
That's not how it works.
But since I don't have all the fine details at my fingertips, I'm going to ask Kevin Reed to give you a full answer. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
As a result, we are going to change how the 2nd part of the process works. Instead of selecting the credit that is closest to its history, we will average the recent average history's for the two computers. It just didn't look like 2 computers could have the same RAC and be working on the same WU and be more than 30% different in crunch time. But I guess it happened. Thanks. You skipped this part: What we found was that there were a few computers that were extremely consistent about claiming very low so they always caused the workunit to check the recent average history Given that the algorithm determined that it was not here the case for the lower, the former part was applied (aka outlier rule): If the two claimed credit values are further then 30% apart, then the code looks at a field in the database which stores the recent average credit granted per second for each computer. Whichever computer's claimed credit per second for the workunit is closer to their recent average credit granted per second has its claimed credit used as the credit granted for the workunit. So in the> 30% rule, even if the slower one was off by 0.001 per second, if the faster was 0.0005 per second off, that's the one used. Couldhave been the other way around: Here's a sample from my RS page: (posted this before) E000000_ 014A_ 00000000e_ 0-- Valid 5-12-08 17:15:37 6-12-08 06:41:13 12.66 127.0 / 127.0 < Moi E000000_ 014A_ 00000000e_ 1-- Valid 5-12-08 16:46:57 5-12-08 22:50:05 3.46 54.9 / 127.0 < Bad underclaimer
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
So my computer took a longer time to crunch than usual (outlier) so we used the claimed credit from the other system, which is why the 2 results are the same, not because they were averaged. Thanks.
|
||
|
|
|