| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Discovering Dengue Drugs - Together
Thread: DDDT Single Redundancy Change |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 87
|
|
| Author |
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Don't think so. Work of the rush/repair/make up type anyway has short deadlines. If the devices is not presently considered a quick returner you'd not even get it.
----------------------------------------Here's a short production table for the last 6 days: Jul.22-08 33464 Jul.23-08 31644 Jul.24-08 31507 Jul.25-08 26112 (day when last 11,000 were send out) Jul.26-08 17124 Jul.27-08 7500 (morning data)
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jul 27, 2008 3:04:36 PM] |
|||
|
||||
|
bieberj
Senior Cruncher United States Joined: Dec 2, 2004 Post Count: 406 Status: Offline Project Badges: 
|
Was wondering why the initlal replication of 3 on the three tasks I currently have. But I see that this is part of the procedure to get the site ready for single redundancy task.
Two in progress, 1 waiting to start. Of course they will be done by tonight.  |
|||
|
||||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Don't know if the bug is still active or not, but the initial replication value was assuming the total copies send out, initially plus the later repair jobs. Look at the distribution times if it was a forced additional or really a initial 3 work unit distro.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
|||
|
||||
|
jonathandl
Advanced Cruncher Joined: Nov 12, 2007 Post Count: 106 Status: Offline Project Badges:
|
You wrote:
----------------------------------------Don't think so. Work of the rush/repair/make up type anyway has short deadlines. If the devices is not presently considered a quick returner you'd not even get it. Then I assume that I have nothing to worry about? Assuming, though, that I do get a job, is the date that the single-redundancy jobs go out already pre-determined, or does it depend on when the last 6.05 jobs get returned? (Now, what would be really interesting is if a very substantial portion of the 7500 represented jobs that already had 2 returns but with inconclusive validation.  )[Edit 2 times, last edit by jonathandl at Jul 27, 2008 7:59:58 PM] |
|||
|
||||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
The Statistics only show Validated jobs.
---------------------------------------- As earlier explained it's "hands on" and estimations are based on past experience. Nothing written in stone and surely tomorrow or the day after the will be an update. Now if you worry about getting a rush job and not finishing in time the option is to switch DDDT out of the My Projects profile until the definitive announcement. That way your part time crunching wont cause it to sit on a half completed job.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
jonathandl
Advanced Cruncher Joined: Nov 12, 2007 Post Count: 106 Status: Offline Project Badges:
|
You wrote:
----------------------------------------Now if you worry about getting a rush job and not finishing in time the option is to switch DDDT out of the My Projects profile until the definitive announcement. That way your part time crunching wont cause it to sit on a half completed job. On the contrary, I want to contribute, and if my crunching results in a net benefit to the project then I'll do it. And I'll keep the computer turned on this time, even increasing the memory and CPU utilization levels. It's not the credit's that I'm worried about. I want to assist in getting the job done. [Edit 2 times, last edit by jonathandl at Jul 27, 2008 9:44:54 PM] |
||
|
|
jonathandl
Advanced Cruncher Joined: Nov 12, 2007 Post Count: 106 Status: Offline Project Badges:
|
Sekerob wrote:
----------------------------------------Don't think so. Work of the rush/repair/make up type anyway has short deadlines. If the devices is not presently considered a quick returner you'd not even get it. I presume the last batch has gone out? I logged in shortly after 9:00 a.m. and got no work – even hit "update" once or twice; therefore, I assume that I am indeed not considered to be a quick returner. Thanks. [Edit 2 times, last edit by jonathandl at Jul 28, 2008 3:41:22 PM] |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sekerob,
do you have any info on the quantity of these that are still in the queue? TIA Here's a short production table for the last 6 days: Jul.22-08 33464 Jul.23-08 31644 Jul.24-08 31507 Jul.25-08 26112 (day when last 11,000 were send out) Jul.26-08 17124 Jul.27-08 7500 (morning data) |
|||
|
||||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Give me a second to wet the finger, run my charting script and I'll get back.
----------------------------------------Jonathandl, the last batch of 11,000 was depleted on Friday Night. We did 32-33 thousand a day before this retooling, so it only takes on a rolling basis about 8 hours to deplete a batch. The beauty I noticed on 2 of my last DDDT jobs was that the extra copy was remotely canceled by WCG, the moment the quorum 2 was complete. Remotely canceled means, WCG readies an instruction which upon first client initiated contact gets transmitted. This was also how the bad FAAH jobs were taken out. The client shows in message log "aborted by project". Maximum efficiency. ciao
WCG
Please help to make the Forums an enjoyable experience for All! |
|||
|
||||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Looks like WCG had a Whoa Nelly 12 hour morning with a total of 106 CPU years of validated work. For DDDT the updated result counts are these, with in this morning session more than 1 batch cleared out.
----------------------------------------Jul.22-08 33464 Jul.23-08 31644 Jul.24-08 31507 Jul.25-08 26112 Jul.26-08 17124 Jul.27-08 27301 Jul.28-08 19807 (12 hour production) If things went to plan, knreed forced a residual rush (likely around 12-13 UTC on Monday). That was for me about 3 hours ago. And a number to complete I don't have. Not much I think. For all we know there are e.g. 7,000 in pending validation waiting on 1 of 2 copies out in the field. Considering completion times and high priority, they should for a majority be back before end of this crunching day.
WCG
Please help to make the Forums an enjoyable experience for All! |
|||
|
||||
|
|