| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 11
|
|
| Author |
|
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges:
|
Saw this report this morning. Was this because first job was 6.03 instead of 6.04?
faah4015_ ALP119911_ xmd02780_ 01_ 3-- Valid 05/04/2008 05:40:15 05/05/2008 05:40:19 11.98 140.1 / 150.7 faah4015_ ALP119911_ xmd02780_ 01_ 2-- Valid 05/03/2008 07:39:28 05/04/2008 05:33:39 21.26 161.4 / 150.7 faah4015_ ALP119911_ xmd02780_ 01_ 1-- Invalid 04/23/2008 07:17:55 04/25/2008 11:19:01 13.78 165.5 / 75.4 faah4015_ ALP119911_ xmd02780_ 01_ 0-- No Reply 04/23/2008 07:14:40 05/03/2008 07:14:40 0.00 0.0 / 0.0 |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
"Are you sure we are ready for eliminating redundant jobs?" Not sure if one has anything to do with the other. The zero redundancy jobs have verification algorithms. At the end of the day if promising results are found, they will be rerun in the lab for verification. Why use redundancy on 99.999% of the tests if they don't end in a promising result. The tests though need to be run to be able to determine if they are or if they are not.
----------------------------------------That 6.03 and 6.04 were not compatible I'd not expected, but a single bit difference is enough to not make them pass the equality test and that could be if e.g. a change is made to the none essential graphics.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges:
|
Sekerob,
Thanks for clearing up the zero redundancy strategy. I was not aware that the intent was to not have redundancy if the the task did not end up with a promising result that is worth further investigation. It now makes sense. Josh |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I don't think that accurately describes it.
The difference with the zero redundancy validator is that it would know which result was invalid without comparing it to another result. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Simpleton understanding: The jobs get a call it a 'quality control' algorithm with it. Once the large data set has been collected, they only would have to take the top performers and meantime watch for indicators that could suggest too many anomalous results either up or down. As I understand it, only HCC is not fit for the approach as for that they need absolute results.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges:
|
Seeing another set of inconclusive results - this time three of them.
faah4027_ Ins_ 007223c_ xmd16490_ 00_ 3-- In Progress 05/06/2008 03:18:14 05/08/2008 03:18:14 0.00 0.0 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 2-- Inconclusive 05/05/2008 08:37:30 05/06/2008 03:15:26 8.45 98.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 1-- Inconclusive 04/26/2008 01:05:01 05/05/2008 08:31:59 10.72 112.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 0-- Inconclusive 04/26/2008 00:54:42 04/26/2008 18:47:59 6.83 131.2 / 0.0 |
||
|
|
BKraayev
Advanced Cruncher Joined: Mar 23, 2005 Post Count: 51 Status: Offline Project Badges:
|
My first FAAH after the upgrade turned out INVALID as well
----------------------------------------faah4001_ ALP007223aPDF_ ISmin_ xmd14690_ 01_ 4-- Error 05/05/2008 19:55:31 05/06/2008 01:05:09 0.00 0.0 / 0.0 faah4001_ ALP007223aPDF_ ISmin_ xmd14690_ 01_ 3-- Valid 05/03/2008 19:52:30 05/06/2008 00:25:29 7.42 109.7 / 116.1 faah4001_ ALP007223aPDF_ ISmin_ xmd14690_ 01_ 2-- Valid 05/02/2008 20:03:54 05/03/2008 19:09:46 7.42 122.4 / 116.1 faah4001_ ALP007223aPDF_ ISmin_ xmd14690_ 01_ 0-- No Reply 04/22/2008 19:59:54 05/02/2008 19:59:54 0.00 0.0 / 0.0 faah4001_ ALP007223aPDF_ ISmin_ xmd14690_ 01_ 1-- Invalid 04/22/2008 19:53:20 04/23/2008 10:40:58 13.54 160.7 / 58.0 ![]() |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
It's unfortunate, but whenever there is a 'No Reply' or an 'Error on the side of any 2 in a quorum right in the middle of a version upgrade, the results are not a perfect match. Here's todays comment from knreed:
----------------------------------------Yes - there is a output file mismatch. We are ready to release faah 6.05 but it will also have a slight difference. I'm doing some work to minimize the impact. Slowing everything down to a crawl doesn't work real well so we are going to try something different this time. Once the zero redundancy validation code has been fully tested this will become a thing of the semi past. Of present and upcoming projects only HCC will retain the standard quorum 2 bit for bit matching. Sorry for any inconvenience. (of course it is against our common sense to blame the 'No Reply/Error' clients. Always consider force majeure kept them away from delivering ;-)
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at May 7, 2008 7:29:09 AM] |
||
|
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges:
|
Seeing another set of inconclusive results - this time three of them. faah4027_ Ins_ 007223c_ xmd16490_ 00_ 3-- In Progress 05/06/2008 03:18:14 05/08/2008 03:18:14 0.00 0.0 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 2-- Inconclusive 05/05/2008 08:37:30 05/06/2008 03:15:26 8.45 98.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 1-- Inconclusive 04/26/2008 01:05:01 05/05/2008 08:31:59 10.72 112.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 0-- Inconclusive 04/26/2008 00:54:42 04/26/2008 18:47:59 6.83 131.2 / 0.0 I just don't believe it. Still inconclusive. faah4027_ Ins_ 007223c_ xmd16490_ 00_ 4-- In Progress 05/07/2008 00:25:06 05/09/2008 00:25:06 0.00 0.0 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 3-- Inconclusive 05/06/2008 03:18:14 05/07/2008 00:20:45 5.68 88.3 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 2-- Inconclusive 05/05/2008 08:37:30 05/06/2008 03:15:26 8.45 98.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 1-- Inconclusive 04/26/2008 01:05:01 05/05/2008 08:31:59 10.72 112.8 / 0.0 faah4027_ Ins_ 007223c_ xmd16490_ 00_ 0-- Inconclusive 04/26/2008 00:54:42 04/26/2008 18:47:59 6.83 131.2 / 0.0 |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
Discouraging situation !
---------------------------------------- faah4024_ ALP402959_ min_ xmd04600_ 01_ 1-- fromveur Invalid 04/25/2008 07:25:24 04/29/2008 08:20:24 7.49 135.4 / 36.9 Cheers, |
||
|
|
|