Posts by devn

1) Message boards : Cafe Rosetta : Team Thread (Ads only, not for discussions) (Message 9834)
Posted 25 Jan 2006 by devn
Post:

Team CF at CPDN, only 32 members but we hold the 3rd spot! If you're not a member of a team, JOIN US, and make your credits count! all crunchers are welcome.

:)

2) Message boards : Cafe Rosetta : Teammember recruiting thread (No. 1) (Message 9083)
Posted 15 Jan 2006 by devn
Post:
team CF, exclusively at CPDN, check it out! with only 31 members, we've got the 3rd spot! if you're not a member of a team, JOIN NOW and make your credits count! :)
3) Message boards : Number crunching : Computational Error (Message 1182)
Posted 9 Oct 2005 by devn
Post:
Well if you have a HT CPU and run it only as 1 instead of 2 your giving up around 15% to 25 % of your Performance Crunching the WU's, may as well Crunch for a Project thats not having a Problem with the HT CPU's.

But that can be hard to do also since it seems all the Projects have some sort of Problem with them. Almost any Project I've run WU's for has a Problem from time to time starting up the Second or next WU when 1 of the 2 running finishes. It's a on again off again thing but all the Projects have the problem. Who knows ... :) [/quote]


i wouldn't mind giving up a little performance if using 1 cpu had worked and allowed rosetta to run w/out errors. haven't had problems with HT and other projects so far.
4) Message boards : Number crunching : Computational Error (Message 1171)
Posted 9 Oct 2005 by devn
Post:
You can run 2 CPU's with the HT CPU's & overcome the "unrecoverable error" by simply Suspending the Rosetta Project & doing a Manual BenchMark.

Do this and mark down when you did it & then just do it again before 5 Days are up when the Server will ask for a Benchmark. I know this isn't probably practical for people with a Ton of Computers but for those with not so many it's a simple work around the Error until the Dev's can figure out why it's happening ...



wanted to try the 1 cpu idea since it had been noted that those computers with 1 cpu weren't having the problem. it didn't work for me but sTrey had success with it.
5) Message boards : Number crunching : Computational Error (Message 1160)
Posted 9 Oct 2005 by devn
Post:

So it seems the benchmarking issue is isolated to multiple CPU machines, whether logical (HT) or physical (dual core/dual processors)


For what it's worth: I've crunched only 17 wu's but haven't had any problems yet and have gone through a couple of benchmarks. I'm running it on a HT cpu, but I've told it to limit cpus to one, since that keeps the cpu temp reasonable.
(& it's set to keep projects in memory)
Hardware is P4 3.2Ghz, 1G ram, running XP Pro

Just another data point.




i have HT but have also tried setting rosetta to use 1 cpu to see if it would make a difference. auto benchmarks caused "unrecoverable error" on a wu yesterday. rosetta is set to remain in memory when preempted but auto benchmarks throws it out of memory.
6) Message boards : Number crunching : Client Errors (Message 1083)
Posted 7 Oct 2005 by devn
Post:
update to previous post: benchmarks just ran and caused "unrecoverable error." why this time and not last, have no idea, nothing else going on.
7) Message boards : Number crunching : Client Errors (Message 1080)
Posted 7 Oct 2005 by devn
Post:
i'm having far more success with wus since upgrading to cc4.72; also, now running only 1 cpu with rosetta even though i have HT. benchmarks no longer cause a problem either.
8) Message boards : Number crunching : A question to the BOINC community... (Message 880)
Posted 1 Oct 2005 by devn
Post:
[quote]I am not sure anyone has really answered the question asked.

First and foremost it should be in the rules and policies page. Donations of equipment and CPU time on, or in, installations are only for approved applications and No means NO. Ask once and that is all. I try to make that clear in the Wiki, but it may have gotten "lost" in the prose ...(snip)




and add it to the homepage, as well, for emphasis.
9) Message boards : Rosetta@home Science : How Is Rosetta Different Than Folding@home? (Message 550)
Posted 26 Sep 2005 by devn
Post:
[Although the growth of the project over the last week has been amazing, our efforts to predict protein structures are still limited by total computing power. Does anybody have suggestions on how to recruit more participants?

thanks,

David[/quote]



post exactly what you said here in the other forums. also, post on the homepage for those people that never read the boards. ask crunchers if they would be willing to concentrate more on your project, at least until you build up a larger user base. you have to let people know what you need.

btw, the optimized app is terrific!


10) Message boards : Cafe Rosetta : Congrats to Travis Tuck and PCZ (Message 437)
Posted 25 Sep 2005 by devn
Post:
way to go travis tuck & pcz!
11) Message boards : Number crunching : If There Is No Screensaver........ (Message 436)
Posted 25 Sep 2005 by devn
Post:
screensavers don't interest me, but after the project's up and running well, it might attract more crunchers if they add one. people pick projects for different reasons, so if it helps, why not?
12) Message boards : Number crunching : My experience with Rosetta WU's (Message 352)
Posted 23 Sep 2005 by devn
Post:
decided to try again. suspended my other 2 projects and started rosetta. success! finished in 8+ hours. ;)



with the optimized app (4.76), my next wu finished in 1hr 38min. :)

13) Message boards : Number crunching : My experience with Rosetta WU's (Message 323)
Posted 22 Sep 2005 by devn
Post:
decided to try again. suspended my other 2 projects and started rosetta. success! finished in 8+ hours. ;)
14) Message boards : Cafe Rosetta : RDC is UOTD (Message 301)
Posted 22 Sep 2005 by devn
Post:
congrats!
15) Message boards : Number crunching : My experience with Rosetta WU's (Message 147)
Posted 18 Sep 2005 by devn
Post:
update to my previous post(ID77). my wu's progress was at 67% but somehow went back to 60% overnight. no idea what happened as my other projects (predictor & lhc) are running normally. aborted the wu; will try at another time. :(
16) Message boards : Number crunching : Warning: Don't shut down BOINC Manager..!! (Message 120)
Posted 17 Sep 2005 by devn
Post:
re previous post...same thing happened to me. i shut down because of powerful t/storms overhead and when i rebooted, the cpu time started from zero although the wu progress showed 40% after a couple of minutes (same as before shutdown).
17) Message boards : Cafe Rosetta : Where are the BOINC regulars...8) (Message 96)
Posted 17 Sep 2005 by devn
Post:
Hi all! Glad to be here. Good luck with your project, UW! :)
18) Message boards : Number crunching : My experience with Rosetta WU's (Message 77)
Posted 17 Sep 2005 by devn
Post:
started crunching my first wu. completion time at first showed 178+ hours but after 33 minutes, completion time now at approximately 8 hours. also, percent done started showing after running about 25 minutes.






©2024 University of Washington
https://www.bakerlab.org