| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Microbiome Immunity Project
Thread: Re-running of old batches |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 9
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Why are we re-running the older batches? Seems like there are too many to just be filling in missed batches. Didn't see any announcement that there was a problem with the previous runs.
|
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18667 Status: Offline Project Badges: 
|
Could this be the batches that had the Atom Pair errors?
---------------------------------------- |
||
|
|
85930294
Cruncher 中国 Joined: Nov 25, 2006 Post Count: 8 Status: Offline Project Badges: 
|
111
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could this be the batches that had the Atom Pair errors? Too many in my opinion. So far, it's been about 2000 batches so it seems more than normal error re-runs or catching up on missed batches. |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Maybe if you post some result names which you consider to be reruns, I'd be able to reference this to my tool history (just 1220 results), noting, there's a steady increase in lead numbers from 00000001 to 00004432 in my history, so 4432 main groups in 7 weeks. Then there's a gap for 3 weeks of cessation and a resume at 00011046 on oct.08 steadily increasing to 13000, then a kick back to 00002949 to now 00004379, mostly _0 and an occasional _1 (sort of 1400 main groups in 2 weeks getting to where tasks were getting through on or before Sep 16.). Don't immediately see any batch number from Oct.8, that was visited by my comps before Sep.16. In all if this there's a single 00007393 and 00009371, but both are _1. Certainly looks like large leaps/(skips?) were made from 00004400 to 00011000.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
To many to post... It was the jump back to 2949 and since, that has raised the question for me. I though we had already processed batches in the range of 2900 to 4581 (latest currently received). in fact, there were errors reported previously in the 4750 to 4850 range that Johnathan had to remove. I had noticed "smaller" regressions in batch numbers but it was easy to see they were picking up skipped batches. I have a hard time believing we missed 2000 batches. Maybe we did and I never noticed....
|
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Sorting on task name, then sent time, not caught any duplicate batch numbers being issued, but then my history is quite limited. If you had them with _0 showing up both in September and October, then the query seems warranted as to what happened... a wonk in the work generation and distribution logic?
MIP1_00002581_0334_0 05-09-17 18:52 MIP1_00002595_0976_0 05-09-17 20:13 MIP1_00002595_1208_0 05-09-17 20:15 MIP1_00002612_1160_0 05-09-17 21:34 MIP1_00002623_0440_1 13-10-17 07:45 MIP1_00002641_0056_0 12-10-17 17:10 MIP1_00002644_0142_0 12-10-17 17:17 MIP1_00002652_0456_0 12-10-17 17:27 MIP1_00002664_0944_1 07-09-17 05:50 MIP1_00002667_0352_0 06-09-17 00:44 MIP1_00002671_0439_0 12-10-17 18:18 |
||
|
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges: 
|
There are some gaps in the batch numbers sent from the researchers. I think the researchers skipped some batch numbers that were not a priority. Likely this is what you are seeing but if you give me batch numbers you think are duplicate I will double check the database. I will also add that our load scripts will not load a duplicate batch name if it has already been loaded before.
Thanks, armstrdj |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for the information.. It is plausible that there are not any duplicates. My history doesn't go back far enough to conclusively determine duplicity. It just seemed that the current running batch numbers were relatively consecutive for a large range (>2000). If these are skipped batches, then they must be very frequent small gaps in the number range that leads one to believe they are consecutive. As far as the load script goes, I assume you have a method to delete batches and re-run them in case of errors. My first thought were these were being re-run due to some identified error.
|
||
|
|
|