| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Human Proteome Folding - Phase 2
Thread: HPF2 Off my list of Projects |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 32
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
My primary installation (strongest machine) has spent the last 117 hours crunching HPF2, with no results, and no points, just burning up the CPU as fast as it can using 50-60% capacity. I think, since it was launched, I've managed to get one result.
I'm going to be using the Grid for something more productive from now on. |
||
|
|
olympic
Senior Cruncher Joined: Jun 12, 2005 Post Count: 156 Status: Offline |
Yeah me too, I'll only be crunching FAAH for a week or 2. I'm still getting invalid results from work units sent out as recently as yesterday. And the number of inconclusives are growing, some in my profile have been crunched 8 or 9 times.
----------------------------------------I'm not upset and I know the techs will get the bugs worked out soon but I want my CPU time to turn in useful scientific results.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I had one yesterday that had been crunched 9 times, I was the 10th cruncher. I was going to pull the pin too but it got Valid status from that 10th crunch. All 10 crunchers got credit, even those with Invalid status.
----------------------------------------Those with 8 or 9 invaid crunches have had 4 or 5 invalid crunches using the buggy Rosetta 5.06. That changes the validation requirement for that WU, the quorum of 3 is insuffucient, 4 or 5 matches are required in those special cases. Note that the quorum of 3 has NOT changed for ALL work units, just for the few work units that meet the condition described above. [Edit 1 times, last edit by Former Member at Jul 4, 2006 2:53:22 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
My primary installation (strongest machine) has spent the last 117 hours crunching HPF2, with no results, and no points, just burning up the CPU as fast as it can using 50-60% capacity. I think, since it was launched, I've managed to get one result. i think you need to give more detail.I'm going to be using the Grid for something more productive from now on. are you using boinc or the ud client? what speed is your pc and how much ram do you have? do you see the graphics in the 'i' screen? explain the last 117 hours - is all of that for a HPF-2 unit you are now working on? |
||
|
|
olympic
Senior Cruncher Joined: Jun 12, 2005 Post Count: 156 Status: Offline |
I understand the "quorum of 3 or majority" rule and would have no problem crunching a WU that has been crunched many times before. But I have a few invalid results from WU's processed with Rosetta 5.07 on July 2. I was surprised to see them since everything was going smoothly since the release of 5.07 back on the 28th.
----------------------------------------Currently I have about 15 inconclusives awaiting validation, so I'm going to crunch only FAAH for a while until all the inconclusive HPF2's work their way through the system and see what happens with them. If all goes well I'll turn HPF2 back on. I appreciate getting the points, no doubt about that(I'm just about to cross the 1,000,000 mark! :) ), but it's no use wasting CPU time turning out invalid results. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Early congrats to you on hitting 1 mil !!
----------------------------------------I've been wondering why they don't erase all the inconclusives from those WUs and restart them at 0. Perhaps there is no automated way to do that. Maybe the only way is to run them through the grinder like they're doing now. I dunno, just guessing but I think if there were any better way to do it then they would. [Edit 1 times, last edit by Former Member at Jul 5, 2006 8:08:28 AM] |
||
|
|
watzkej
Cruncher Joined: Mar 12, 2005 Post Count: 12 Status: Offline |
That's really strange that others are having issues with the client. I have 5 machines: 3 running UD client and 2 running BOINC. I've only had 1 Error result in all of the units that have been crunched by the 2 BOINC clients. So it doesn't seem the project is all that bad. I'm using the latest BOINC client, 5.4.9.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi watzkej,
There are still some bugs in Rosetta 5.07. Nothing like the bugs in 5.06, but the staff are picking out problem work units from the many that we run and are trying to debug to find where the errors are. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Rosetta 5.07 is a vast improvement over 5.06. The reason the techs were able to make those improvements is because dedicated crunchers stayed on task and didn't give up on HPF2.
----------------------------------------The faster we crunch and re-crunch problem WUs the sooner the WCG techs will find the bugs and squash them. Remember, we have 2 jobs to do here: 1) improving the science of protein folding 2) improving the science of grid computing Crunching the WUs gives the techs valuable clues. From those clues they will find more bugs and develop the next version of Rosetta which will be even better than version 5.07. Contrary to what scubaboy and other people have said in this thread, the sensible and helpful thing to do is to keep on crunching HPF2. The WUs with a high number of Inconclusives will work their way through the system faster if we all keep on crunching HPF2. [Edit 2 times, last edit by Former Member at Jul 5, 2006 6:25:44 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I've seen several references to version 5.06 & 5.07 of the Rosetta Pase 2 software. I was under the impression that my software would automatically update when a new WU was downloaded, however mine continues to run version 5.05.03. Is there a way for me to get the updated software or is it even necessary? Thanks. Jim
|
||
|
|
|