| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Human Proteome Folding
Thread: Curious about those HPF problems |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 10
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
will we be told what the problems were with some recent work units that got removed?
----------------------------------------how are the current units different from the problematic ones? and, were the the non-BOINC work units still good? i do not run BOINC. what would be the difference? thanks. [Edit 2 times, last edit by Former Member at Jun 13, 2006 7:19:32 AM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
curiousity kills the cat ...... personally, the faulty batches were taken off line as i read in a knreed message.....i'm sure they've diagnosed how they were different in compilation, thus I let it be and crunch FAAH in meantime so my machine does not sit around idle. I've gotten a few Rosetta's from different batches in meantime which were perfectly fine.
----------------------------------------ciao
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jun 5, 2006 5:43:49 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
bump............
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
halfcard, no bump at all. As for UD versus BOINC, the material is same but slightly differently compiled i.e. you cant take a BOINC WU and put it in UD agent v.v. I think i read that possible a UD batch same time as BOINC from that Hxxxxx series ???? went belly up and was removed. That suggests that the data set itself might have had issues prior to compilation for crunching. Way over my head, i just crunch and interject with comments.....just keep going, we cant get enough of your cpu time
---------------------------------------- PS for the moment i'm only getting GUxxxx series Rosetta of different lengths/duration....they're all fine too 
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Jun 13, 2006 10:26:15 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
sek, i agree with you in principle. but...
i am still bothered by the UD experience i had, where the managers decided to run old work units again without telling anyone, apparently to conceal problems they were having! |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
I think that the ethics under which the WCG projects are run and the "partners" accountablility is beyond a shadow of a doubt. Never will WCG present double data just from stopping anyone to leave...if ever, i'd take personal afront and will voice this as i did at UD.
----------------------------------------Of course, if an improved way is found to recrunch promissing data sets to even better determine hits, thats fine with me. In analogy, in Geoscience, Seismic companies collected lots of data & curves....as the modelling technology improved, recrunches brought many carbon holding zones to light previously completely missed. I'm sure a "board" will review this and will be advised in member news should the situation occur. ciao
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 3 times, last edit by Sekerob at Jun 20, 2006 11:40:22 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi halfcard,
The WCG is trying to get corporations and educational institutions to donate their spare CPU cycles to charity. These institutions have zero (0) interest in points. Running old work units for points might appease some individuals who only care about scores, but would disgust those institutions who support the WCG with their bandwidth and electricity. So it will not happen. If we cannot pick up any new work units, we will go into a frenzied panic behind the scenes while the clients begin phoning home daily to look for any new work.  The problems with bad batches are normally caused by improper preparation of the headers that have to be attached to each work unit. That can be fixed by reinstalling new headers on the work units. It happens.  Lawrence |
||
|
|
Viktors
Former World Community Grid Tech Joined: Sep 20, 2004 Post Count: 653 Status: Offline Project Badges: 
|
The problem work units mentioned before had a subtle input data problem which was making Rosetta quit. These were corrected and run successfully without incident. I thought I posted something elsewhere on this, but perhaps not prominently enough. Sorry. We have been heads-down working on getting several new projects ready. We are actually very very near to completion of HPF1 and there should be an announcement soon.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for re-posting that info, Viktors. With the information overload that many of us experience it's not surprising we forget. Some days I think even if you posted it on a giant billboard with flashing lights on my lawn it still would not sink in <sigh>.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for all your responses and I am glad the problems were figured out and corrected.
|
||
|
|
|