| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: New DSFL Beta |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 38
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Except the valids and pendings, I also returned 20 of the known error batches. Will this influence the reliability of my hosts and lead to not sending repairs to that hosts? CP, "Only" Invalid results go against reliability i.e. not flat error or aborted. All statuses, including invalid, go against the daily quota and that does get reduced down to 1 per day when there are too many, but you'd have to have very serious issues as a normal running system has a quota of 80 per core, up to 32 cores. New science... I'd like to keep a mix going, so in between, a stable science result can increase the quota again, but 80 per core/day is hard to bust. --//-- |
||
|
|
JSYKES
Senior Cruncher Joined: Apr 28, 2007 Post Count: 206 Status: Offline Project Badges: 
|
I have had 4 random WU's, a 53 that ran instantly to error (fine); a 19, a 24 and a 28 - the latter all ran OK, 19 now V; 24 & 28 in PV. The only odd thing noted was they came with 12hr+ estimated processing time but were all completed in 1.75hrs....was this a random setting on issue? As has happened previously?? Other than that, all OK. Look forward to seeing more!!
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello JSYKES,
Uplinger has said that Beta DSFL work units created to test targets will be shorter than normal. Ignore the time estimates. They seldom have much meaning for Beta units. Lawrence |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Ended up with 9 units on this Beta run (without even trying), the usual suspects errored out immediately, the type 16 ran for 6.7 and 4.3 hours on beefy machines, but everything else ran glitch free and is either pending or validated. (With the exception mentioned in the DSFL thread, that system has yet to produce a valid result on this science.)
----------------------------------------Edit to fix run times.  Distributed computing volunteer since September 27, 2000 [Edit 2 times, last edit by KWSN - A Shrubbery at Sep 9, 2011 2:39:22 AM] |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
I'll take whatever betas you send out. I might hit gold before I take a dirt nap.
----------------------------------------
Crunching for humanity since 2007!
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Type 16 on my slowest machine (2.4GHz Pentium 4, the only single-core I have left), running XP, took just shy of 17 hours to finish; Type 52's on one of my fastest quads (3GHz Phenom II) errored out in about 2 seconds.
I had to turn off the (non-beta) DSFL on my Core 2 Quad Q6600, because all it seemed to be getting was type 40s, which were running about 12% efficiency, eating nearly 10 hours of runtime while producing only 1 hour of CPU time. |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1406 Status: Offline Project Badges: 
|
"Only" Invalid results go against reliability i.e. not flat error or aborted. Thanks Sek for refreshing my knowledge. It must be hidden in one of my grey chambers. I repeat to myself for better memorizing: "Only invalids decrease the daily quota" All statuses, including invalid, go against the daily quota and that does get reduced down to 1 per day when there are too many, but you'd have to have very serious issues as a normal running system has a quota of 80 per core, up to 32 cores.  New science... I'd like to keep a mix going, so in between, a stable science result can increase the quota again, but 80 per core/day is hard to bust. I'm mixing the BETA's with simaps, but although stable science, I'm afraid that doesn't count. I will add DSFL to my mix again, when simap september work is done. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
"Only" Invalid results go against reliability i.e. not flat error or aborted. Thanks Sek for refreshing my knowledge. It must be hidden in one of my grey chambers. I repeat to myself for better memorizing: "Only invalids decrease the daily quota" All statuses, including invalid, go against the daily quota and that does get reduced down to 1 per day when there are too many, but you'd have to have very serious issues as a normal running system has a quota of 80 per core, up to 32 cores. New science... I'd like to keep a mix going, so in between, a stable science result can increase the quota again, but 80 per core/day is hard to bust. I'm mixing the BETA's with simaps, but although stable science, I'm afraid that doesn't count. I will add DSFL to my mix again, when simap september work is done.There's something(s) incorrect with that bolded line, so plz do not repeat in sleep either ;O) --//-- |
||
|
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3716 Status: Offline Project Badges:
|
It'd be nice if new betas could be posted in the beta announcement section for those of us who have a watch set on that section to save needing to check the forums a few times a day for info about new tests. widdershins ,  With so small beta sets, when you receive the forum watch email all beta WUs will have already been distributed for hours. ---------------------------------------- [Edit 1 times, last edit by JmBoullier at Sep 9, 2011 8:21:15 AM] |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1406 Status: Offline Project Badges:
|
Thanks Sek for refreshing my knowledge. It must be hidden in one of my grey chambers. I repeat to myself for better memorizing: "Only invalids decrease the daily quota" You opened my eyes: Only invalids decrease the daily maximum quota.There's something(s) incorrect with that bolded line, so plz do not repeat in sleep either ;O) Better? |
||
|
|
|