| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Discovering Dengue Drugs - Together
Thread: Inconclusive Result |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 6
|
|
| Author |
|
|
martianmoons
Cruncher USA Joined: Nov 29, 2006 Post Count: 49 Status: Offline Project Badges: 
|
You may wish to check out the below workunit which shows 2 Inconclusives and 1 In Progress.
Since 2 results are needed for a quorum, the third person working on this is wasting their time since the workunit cannot be validated by their one result. And something may be wrong with the workunit in any case. Thank you. Created: 09/08/2007 01:56:52 Name: dddt0101a0049_ZINC01352742-0000_02 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
No, that's not how it works. The third result will be compared to the two inconclusive results, and one will be changed to valid and the other to invalid.
What you see is perfectly normal. |
||
|
|
martianmoons
Cruncher USA Joined: Nov 29, 2006 Post Count: 49 Status: Offline Project Badges:
|
No, that's not how it works. The third result will be compared to the two inconclusive results, and one will be changed to valid and the other to invalid. What you see is perfectly normal. Thanks for the heads up! Does that mean that inclusive results may source from two different processor types that yield slightly different floating point results? |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
per knreed, bit for bit equality is required within the homogeneous redundancy rules WCG applies.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
No, that's not how it works. The third result will be compared to the two inconclusive results, and one will be changed to valid and the other to invalid. What you see is perfectly normal. Thanks for the heads up! Does that mean that inclusive results may source from two different processor types that yield slightly different floating point results? No - the invalid result that will be identified will be invalid due to something being computed wrong on that computer - likely due to overclocking or a memory problem. Any differences in computation due to differences in floating point results are handled by our Homogeneous redundancy policy which only sends the workunit to computers who will compute the same result. |
||
|
|
martianmoons
Cruncher USA Joined: Nov 29, 2006 Post Count: 49 Status: Offline Project Badges:
|
...No - the invalid result that will be identified will be invalid due to something being computed wrong on that computer - likely due to overclocking or a memory problem. Any differences in computation due to differences in floating point results are handled by our Homogeneous redundancy policy which only sends the workunit to computers who will compute the same result. Thanks! This is the first error I have ever gotten on BOINC on my slightly overclocked AMD Athlon machine. Perhaps the application is very sensitive to overclocking? Or maybe there was a communication glitch? At any rate, I will keep an eye on things. I was just surprised given my excellent track record on many other projects! ;) |
||
|
|
|