Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
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 |
I believe the major problems are confined to people with 8 cores. 4 cores may also cause problems, and hyperthreading may be a problem (but I don't recommend hyperthreading anyway, and this is a good example of why not). It depends how much memory you have available.
----------------------------------------If in doubt, check your page fault count. As a rule of thumb, I would say "don't run HCC on more than 2 cores at once". Also, "allow 1GB per core running HCC". Like all rules of thumb, these may be totally useless. But I believe they cover the vast majority of cases reported here. edit: Sekerob reminds me that L2 cache is one of the (if not the sole) limiting factor. [Edit 1 times, last edit by Former Member at Jan 24, 2008 5:23:04 PM] |
||
|
courine
Master Cruncher Capt., Team In2My.Net Cmd. HQ: San Francisco Joined: Apr 26, 2007 Post Count: 1794 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Didactylos: A very few people are still trying to blow this problem out of proportion. Only a few hosts with a particular set of characteristics are affected significantly. A larger group of hosts see a very minor effect. 1/2 my crew is effected. If I did nothing to stop the problem, it kills about 1/10 the team score. In the global statistics, this problem simply isn't visible. I am convinced it has no relevance to leagues, either. How could it? The answer is just 1 post back D. And for anyone who does observe that any of their computers are significantly affected by this, I urge you to switch to a different project for now. BINGO! About a month late, but BINGO!!!! Now from my team thread, Nov 30: OK, I'm wondering how things are going in regards to HCC work units. What we need to find out is what kind of machines crunch this work unit effectively. I believe that cargod is running ok, but a few of us may be slowing way down running HCC. This is OK if you understand this is happening and consent, but if you are not sure and have seen a dip in your points stats since the project began, then please feel free to steer clear until the HCC project has broken in a bit more. The 2 categories of systems that I have seen most at risk: the slower than 1ghz and multicore machines. This doesn’t mean you are at risk of you machine, just that the crunching may seem slower or running inefficiently. This note is not a directive, but an advisory. Many of us can stand the rough waters that DC is known for. For the new member though, a wrong wave can find them overboard. So, like any good ships captain, I’m just alerting you to a stormy sea. Now about this: Lavaflow: huh, was it not Didactylos that wrote that January was debugging month. Maybe suggest to WCG to reintroduce CPU minimum requirements for Conquer Cancer and up front shut out anyone not meeting the optimal system specifications? If it's not solvable it might be the only option left, but would not want speculate how your 'sheep' will react to that. Didactylos: It has been said a few times, now. I'm sorry they escaped your notice. A "Known Issues" post might have helped, but usually they are reserved for more immediate problems. Besides, it's amazing how often posts in "Member News" and "Known Issues" get overlooked... :-) Courine: I have "Known Issues" and "Member News" on watch. If anything there was to be viewed regarding this thread, I would have seen it. Where is the thread you talk about? Nice job at character assassination, but it will only work on the sheep. You may have said something way back when, in this thread. Now go get the quote please from the others.. Not to clear this issue, but so the info can bee seen in the here and now. What I am asking is you "throw down". Not for me, but everyone here. ![]() ![]() ![]() [Edit 5 times, last edit by courine at Jan 24, 2008 7:23:55 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I believe the major problems are confined to people with 8 cores. 4 cores may also cause problems, and hyperthreading may be a problem (but I don't recommend hyperthreading anyway, and this is a good example of why not). It depends how much memory you have available. If in doubt, check your page fault count. As a rule of thumb, I would say "don't run HCC on more than 2 cores at once". Also, "allow 1GB per core running HCC". Like all rules of thumb, these may be totally useless. But I believe they cover the vast majority of cases reported here. edit: Sekerob reminds me that L2 cache is one of the (if not the sole) limiting factor. Actually, I have observed this problem on my old Pentium 4 machine, and on my dual core machine as well. Steve |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Courine, was that post addressed to me?
If so, I have no idea what you are talking about. If, (as I guess) you are asking me to look up specific posts for you, then the answer is no. I'm here to help people, not to get worked up over trivialities. Feel free to look them up yourself. You have made a number of incomprehensible comments in this thread, and frankly I'm a little disturbed by them. Who are you calling "sheep"? Who are you accusing of "character assassination"? What do you mean by "you even tried to infer I was a hacker with in a few hours of my arrival"? We seem to have a communication problem, and I should very much appreciate it if you would clarify your meaning. |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I've noticed the excess paging and poor point scores on several of my systems. My impression from reading the posts were they'd fix it when they got around to it - I just removed myself from the project. My electric bill is high enough as it is without running bad computing.
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks Ady. Just wondering how this would show up in the results of a challenge that isn't broken down by projects? Results, points and runtime are all generic with no distinction of which projects were crunched, or not crunched, to arrive at the numbers. Thanks ![]() Respectfully, Diana G. Ah, thats something you will have to ask courine - I have no idea how these leagues work. I was just clarifying courines post for you ![]() |
||
|
courine
Master Cruncher Capt., Team In2My.Net Cmd. HQ: San Francisco Joined: Apr 26, 2007 Post Count: 1794 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Un-official Grid.org Refugee Transition Center. > http://www.worldcommunitygrid.org/forums/wcg/...thread=12916&offset=0 13th post in
----------------------------------------The topic was about ‘treating’ PTSD in the group from grid.org closing. Courine: YES. We must work quickly! Maybe take him back to the grid forums for a sec, then bring him up SLOWLY! I’m copying some old grid forum pages right now. If we have further cases after tomorrow, we can use them to curb the withdrawal [from the forum at grid.org] Didactylos: Did the web archive get them? edit: nope. Google's not touching it, and the Internet archive says there's a robots.txt exclusion. D, what were you saying? How was I getting a copy of grid.org’s forums? I clarified this later, but why would you comment? This may just be bad communication, but I see this too much with you and not with others. Case in point: “Henny Penny” thread. I’m sure others here could fill you in on that ditty, and how it related to “BOINC Damaging the Environment” thread. That was incredibly telling. Now the problem is I need you to prove your statement: Where is there any real warning that there is a problem. What I see is you down playing like we are making a mountain out of a mole hill. And its me that is doing it. That I am the problem? If you think I’m a problem, you will see others here know I have avoided a problem you don’t acknowledge. What you will find about the leagues is that your team will stand to gain a great deal if leagues here develop. The reason is: my distant history is your future when it comes the challenges. Ask Geo, I’m working with him to get the stats automated enough to run them for months. Making WCG Sports gripping, without the need for WCG to be burdened. ![]() ![]() ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
WHOA!!!
Everyone hit the brakes... I'm interested that Sekerob mentioned that this can also be an L2 Cache problem. Which, if I'm right, is only resolved in the BIOS setup menu. Also, is everyone on the same page as far as running WCG on a separate partition, even if your machine(s) run WCG exclusuvely? It's that doggone kernel that keeps trying to override your operating systems' natural tendency to want to be #1. BOINC, BSD, UNIX and the like operated originally in large server environments, where neither Windows, nor MAC used to tread. I utilized these applications and languages on a Sun Solaris system years back, and the only hassles was when we used the VI interface incorrectly. That was it's text editor. I strongly believe that if one wishes to run all of the available projects, one must be willing to tweak their preferences, both in the BOINC location, as well as their own system environment, to agree. And the more projects one is willing to run, the less robust ones' statistics will be. I am of the strong opinion that HCC is a single core, exclusive run. That means standalone. Because of the way the it behaves as encoded. And I bet if it is run on a separate partition, their may be a turnaround in opinion. The opinion stated here is the opinion of the writer, solely intended to enliven the technical discourse regarding our crunching community. Let's talk shop people! It's about principles, not personalities. Let's be civil with one another. ![]() ![]() ![]() |
||
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2130 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
WHOA!!! . You may have the option to turn the cache on or off in bios but that's it. You have no say in how it's used. Everyone hit the brakes... I'm interested that Sekerob mentioned that this can also be an L2 Cache problem. Which, if I'm right, is only resolved in the BIOS setup menu. Also, is everyone on the same page as far as running WCG on a separate partition, even if your machine(s) run WCG exclusuvely? It's that doggone kernel that keeps trying to override your operating systems' natural tendency to want to be #1. Maybe I've misunderstood you but the kernel is the core of the o/s - see http://en.wikipedia.org/wiki/Kernel_(computer_science).As far as running on another partition. I know there was a thread about that - which i haven't read - but it is beyond me how running it in a different partion - even if its on another drive - will do any good. My boinc directory and all subdirectories are about 150meg for a quad core processor. This will easily fit into 4 gig of memory that I have. Plus, if you keep the defaults, boinc writes to the hard drive only once every 60 seconds. There should be no need for reads since as I said, the entire boinc directory represents about 4% of available ram. But even if I'm wrong on both of those points, unless an application is disk intensive, which boinc is not, disk access is several orders of magnitude slower than memory access anyway. So once the application has to go out to disk, you're already screwed - regardless of where boinc is installed. And the more projects one is willing to run, the less robust ones' statistics will be. I am of the strong opinion that HCC is a single core, exclusive run. That means standalone. Because of the way the it behaves as encoded. And I bet if it is run on a separate partition, their may be a turnaround in opinion. There's no logical reason that this should be true. It may in fact be true, but to the extent it is, the blame lies with the programmers not the users. Let's talk shop people! It's about principles, not personalities. Let's be civil with one another. Some of us have already tried that - maybe you'll have better luck. As long as you agree with with the official rap, I'm sure you won't have any problems. But once you've expressed an opinion critical of WCG or boinc, come back and let us know how that worked out. That's what Courine is talking about when she mentions "sheep" - people who go along with the herd and accept whatever they're fed - regardless of whether or not it squares with reality.![]() ![]() ![]() That's why I didn't even bother responding to Did's post. Anybody who has worked on HCC knows its not limited to 8 core machines or machines that are hyperthreaded or machines that have less than 1gig of ram per core. I have 1, 2 and 4 core machines that ran HCC and I still had the problem even without hyperthreading and at least 1 gig of ram per core. ![]() ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
interesting summary of page replacement algorithms:
http://en.wikipedia.org/wiki/Page_replacement_algorithm |
||
|
|
![]() |