| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Help Conquer Cancer
Thread: [RENAMED] Some concerns regarding the HCC project (page fault and poor performance, in particular (but not only) by multi-core hosts; #cores>2) |
| No member browsing this thread |
|
Thread Status: Locked Total posts in this thread: 210
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Are any of the CA's able to say whether the techs are looking into these issues?
I am reading posts at several forums now where these credit issues are being discussed and many are pulling from the HCC project onto others - I would hate to see the amount of work done on this valuable project decrease. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The techs have some genuinely urgent priorities right now, like getting HCC back working on Linux.
I expect they will address the pagefault issue once the urgent issues are resolved. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
No disrespect D, but is there any chance we can get a comment direct from the techs here about this along with their thoughts? - if folks are abandoning HCC on Windows as a result then it could become an urgent issue.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Maybe one of the techs will comment. I'm not going to bug them about it, though. The HCC/Linux and internal server errors are more important, because they stop people contributing altogether.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I appreciate they are your personal thoughts D, but is it not possible to put them aside for a moment? - you as a CA are our link to WCG - all we ask is that you, or if you're not willing, then another CA, bring this to their attention and ask them to just have a quick look at the thread and perhaps comment so we can see our concerns are being noted and considered.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The page fault issue was noted in another thread long before this one. The techs are aware of the issue. Did I not make that clear? Forgive me.
I see no need to waste their time with a rehashing of the issue when they are already so busy - particularly a thread that is looking at this solely from a credit viewpoint. I prefer to pick my battles, and right now, important though this is, it is trumped by other issues. Remember - it doesn't matter which projects you contribute to - WCG are controlling the balance. So, until this is fixed, contribute to some other projects, and the HCC work you would have done will go to someone else. I know it's more satisfying to work directly on what interests you, but in the long run it makes no difference. WCG have limited resources. You can have the techs posting in here, or you can have them actually fixing the problem. That is the only reason why we don't demand their attention when it won't do any good. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Unfortunately your reply, as is often the case, didn't make it clear - apology accepted
 |
||
|
|
zombie67 [MM]
Senior Cruncher USA Joined: May 26, 2006 Post Count: 228 Status: Offline Project Badges: 
|
I see no need to waste their time with a rehashing of the issue when they are already so busy - particularly a thread that is looking at this solely from a credit viewpoint. FWIW, I was (am) not looking at this as a credit issue. The credit part is only one of the symptoms of the real problem. The credits just help to make it more obvious.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Heheh. A nice theory, but I spotted it without studying anything credit related (as did others). I noticed it when I gave HCC a once-over right after launch, and reported it here http://www.worldcommunitygrid.org/forums/wcg/privateviewthread?thread=16933 on Nov 2 (almost a month ago! this is why I gave this a low priority - nobody else even noticed the problem for another two weeks).
uplinger replied, but not optimistically: 1) This is part of the imaging processing and I'm not sure how much we can actually decrease the page faults. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
 I've been avoiding saying too much in this thread because my views are so different from those being expressed. The really important thing is to knock out the results quickly so that the project scientists can work on converting the output into a set of useful techniques to apply in crystallizing proteins. Optimizing the program for long term use is a waste of time. If this computational technique does become a standard technique, the computers it will run on will be dramatically different from today's 2 and 4 core clunkers. Way to go, Didactylos!  Lawrence |
||
|
|
|