Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 1130
|
![]() |
Author |
|
JSYKES
Senior Cruncher Joined: Apr 28, 2007 Post Count: 201 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I'm pleased to know that someone is receiving WU's as, so far, I've only had 7 WU's in total (very annoying), only 2 'A's, the rest short runs so the prospect of bronze is looking remote.....don't know why I haven't seen any for the last week...hope there might be a better share out of the 22million 'C' WU's in due course!!
----------------------------------------![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I just had to abort a type A, Running on an old 2.8 P4HT that I was running Windows 7RC on. I run it headless so I didn't investigate until a similar XP machine that had also scored two type A's at the same time had returned two yesterday. March 1 Windows 7RC started shutting down every two hours without a save, so because the wu takes so long to checkpoint it wasn't reaching the first checkpoint before the reboot, its been doing this since the 24th. I still have one left that I will see if it has a chance to checkpoint and finish before deadline now that the HT slowdown is removed, I'll know by tomorrow morning. If not I'll have to abort that one too. Too bad wasting all those machine cycles but it didn't show up on normal length stuff, I would just monitor that the machine was returning things on a daily basis in the results screen. Time to change it over to Linux! Hope it goes to somebody with a fast machine. Most of mine are reliable but s-l-o-w compared to today's multicore processers.
|
||
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
When I read thing like this I feel really depressed. First, remember all this is a volunteer thing, no one obliges you to crunch in one or x projects. Do what you want, but does not make pointless demands to the project. What you are actually saying, more at the stage where DDT2 is now, is that you want almost in exclusivity all the units that the project is issuing. Don't you realise the arrogance of that statement?. Don't feel depressed Trotador, you missed my point. I am getting very little from DDDT2 at the moment much less than others but we are in a starting phase and I have no problem with that. On NRW I decided to stop some time ago at 2 years and not push to the 5 years to allow others to get WU's and get a badge before the end of the project. But it was my decision not imposed by the system and this makes a big difference. What I was adressing in my post is the daily quota per thread. That quota is not related to one specific WCG project it is an absolute limit per thread (my understanding). This means that if we go with very short timed WU's (HCC or DDDT2 or else) then once you have gobbled your daily WCG WU quota you just get nothing more. You stop crunching for WCG. This means you have to crunch another BOINC project or whatever other project outside WCG to avoid staying idle. This is what I do not want to be oblidged to. But if it happens, c'est la vie. Now what does this mean: 1) It means first that projects will finish quicker and that is all good for scientists. 2) It means that there is more crunching power available on the grid and this is also good as it will allow scientist to design new projects, or adress a new class of problems. 3) Shorter times per project means a risk for the slower machines that they will not be able to reach the top badges before the project ends. This is a real problem as every contributor, slow, fast, big, small must be rewarded in an acceptable way. I am not sure that to limit is the right way to go because the real effect is to limit the crunching power of the whole grid, and this is the opposite of what we really want. ![]() [Edit 1 times, last edit by Hypernova at Mar 30, 2010 11:46:58 AM] |
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Your elaboration is kind of hypothetical. Presently the single science average run time would have to be on a continuous average at less than 18 minutes per task, wallclock. We've highlighted several times in past weeks that we monitor this closely. As and when the situation would occur, the techs have the ability to reset the quota at an individual level or up it in general, temporarily. WCG has to consider that the larger community has to have access too.
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
HutchNYC
Advanced Cruncher United States Joined: Nov 27, 2005 Post Count: 97 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Why all this concern about the daily quota limit? You're making all these posts about how you think that this is such a bad thing, computers will be sitting idle, etc.... Yet, I haven't seen a single post about anyone hitting that limit and running out of work.
----------------------------------------I'm sure some people have hit the daily download limit, but that was when they were bumping up their cache. It still leaves their machines plenty of work to crunch in the meantime. Until anyone actually has a real world example of this quota limit causing their machine to go idle, I'd suggest not worrying about it. I'm sure the tech's thought about this possibility and set the limit accordingly. |
||
|
HutchNYC
Advanced Cruncher United States Joined: Nov 27, 2005 Post Count: 97 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Looks like Sek agrees and posted quicker. ;)
---------------------------------------- |
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Hutch, this has happened far as i can remember only 2 times, for early days DDDT-1 and maybe FAAH, can't exactly remember. Then the quota was increased from 80 to 120 or 160, as I noted in previous post, temporarily. Upping the quota further above what the top device can hack in 24 hours is not an option... simply stay connected to backfill.
----------------------------------------hmmm, wonder if it actually happens if a rule could be set to allow similar as with Beta ... return result 80 for the core inside the 24 hour frame, valid of course or at least with edit: struck
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Mar 30, 2010 12:57:03 PM] |
||
|
HutchNYC
Advanced Cruncher United States Joined: Nov 27, 2005 Post Count: 97 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Sek, I vaguely remember getting a bunch of WU's with a very short runtime a few years back. I don't remember what project it was either.
----------------------------------------Point being, I doubt that there are many machines where this quota limit will actually be an issue. If it does actually come into play though, I like your idea of setting it similar to what the beta is now. |
||
|
damir1978
Senior Cruncher Joined: Apr 16, 2007 Post Count: 397 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I received only 2 Wu's (A) at the beginning. I was sharing this Project with 3 others and after the A Wu's I got nothing.
----------------------------------------Now I'm only on Dengue Project from more than 2 days and I'm still getting nothing. Is this project sort of Intermittent at this point? When it will get full streamed? |
||
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
damir1978, that last question is really now beyond the point of needing answering. The supply will be more or less throughout the phase non-continuous.
----------------------------------------Et al, frankly if I were the boss, I'd get all the remaining bugs schmertzed out such as the -161 and the occasional 1819, I'm already detecting whine [again pardon my frankness], then build up a good stock of C after getting a couple more A/B type sets worked out and then let it rip, however the cards may fall. Watch it, we just passed 100,000 completed for the project. Not even 10% of the 1.2 million C type there will be, for each of the 18 targets, plus the ts bit.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Mar 30, 2010 1:05:56 PM] |
||
|
|
![]() |