Message boards :
Number crunching :
Report all problems here...........
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 . . . 10 · Next
Author | Message |
---|---|
MadMaC Send message Joined: 4 Apr 01 Posts: 201 Credit: 47,158,217 RAC: 0 |
Whats the story behind these limits - are they going to do away with the cache entirely? Are these limits a test and they will go back up? |
Jim_S Send message Joined: 23 Feb 00 Posts: 4705 Credit: 64,560,357 RAC: 31 |
Whats the story behind these limits - are they going to do away with the cache entirely? You'll find more info on that in Jeff's thread in the NEWS threads. *EDIT*Sorry They are in NC.*/EDIT* I Desire Peace and Justice, Jim Scott (Mod-Ret.) |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
You'll find more info on that in Jeff's thread in the NEWS threads. No, they're hidden. Perhaps that a moderator who can still read the thread can copy the info Jeff posted and put it somewhere for all to read? {hint}. :) |
Link Send message Joined: 18 Sep 03 Posts: 834 Credit: 1,807,369 RAC: 0 |
Any Ghost WUs shown in the task lists probably do not affect the limits. AFAICT the limit logic does not do a database query to find out how many are in progress (that would be expensive like the "resend lost work" feature). So it is based on information in the request, IOW the work the host does know about.Joe If the information is taken from the request, then anyone who like experiments and knows how to edit the client_state could "move out" all (or at least most) of the VLARs and later move them back (after he got some WUs for the GPU). |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13847 Credit: 208,696,464 RAC: 304 |
So, Grant, please don't belittle those of us who have taken this project to another level. I'm not belittling anybody, just pointing out that most of the noise on these forums is neither warranted or justified. Some like you ridicule us for doing so. I only ridicule those that are deserving of it, and even then only repeat offenders. Those that act like adults get treated like adults. We only complain because we want to do more, my friend. And over the years i've found that complaining only leads to resentment & is probably the surest way of not getting what you want. Offering valid criticism, and suggestions of how to resolve a problem (that is when there is an actual problem) are more likely to get the desired result. And then hoefully the maturity to realise this is the real world & while all things are possible, unless there is the funding & personel not everything can be done & what you might consider to be the most pressing issue may not be so for the vast majority of people involved. Grant Darwin NT |
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
Yes, you could say I want my RAC...... "Time is simply the mechanism that keeps everything from happening all at once." |
Jim_S Send message Joined: 23 Feb 00 Posts: 4705 Credit: 64,560,357 RAC: 31 |
You'll find more info on that in Jeff's thread in the NEWS threads. Me thinks I'll leave that call on Jeff. ;-] |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
Honestly, the view I have.. With the seperate limits on CPU and GPU, I see no reason the TOTAL task limit needs maintained. That was the problem with people not getting both last outage, and continues to be a problem after the latest outage. The servers and network survived the initial onslaught. Turning the numbers up slowly as things continue to calm down... And I think we have a winner. [edit]Oh.. and the time calculations on my CPU units are really still WAY off.[/edit] Janice |
Rattledagger Send message Joined: 22 Jan 00 Posts: 23 Credit: 1,010,202 RAC: 0 |
Honestly, the view I have.. Without a total server-side limit, chances are some computers would "suddenly" get 128 cores or something... |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13847 Credit: 208,696,464 RAC: 304 |
With the seperate limits on CPU and GPU, I see no reason the TOTAL task limit needs maintained. My feelings as well. Grant Darwin NT |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
Honestly, the view I have.. 40 per GPU limit. That should cover it. 5 per CPU limit. That should cover it. Janice |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
He does, however, read boinc_alpha, and if you manage to pitch your report in words of one syllable while he's watching, you can get changes made. I managed to trigger him into changeset [trac]changeset:21877[/trac] on Tuesday night, though sod's law dictates that Rom released v6.10.58 during the half-hour between my report and David's fix. God knows when we'll get to test it. But Miep is right to raise the question: why do they spend time tarting up the forums for 'us' to use, while 'they' commune elsewhere in some ivory tower? Aren't we all in this together? |
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
Richard, PM me if you really want some input, but mine is readily chimed upon what others post in the NC. I am sorry that I don't wither away at my thoughts here........ There are few questions that others have raised that I have not been the first one to ask. Sorry if I have not been more civil to you, and caused you to withhold your responses to me. You seem to take me like a jackass these days, and choose not to respond. It is not my intend, my friend. "Time is simply the mechanism that keeps everything from happening all at once." |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
But Miep is right to raise the question: why do they spend time tarting up the forums for 'us' to use, while 'they' commune elsewhere in some ivory tower? Aren't we all in this together? The BOINC developers always said that the forums are for the users, not necessarily for the admins/developers. No matter what it says in the Wiki. ;-) But in essence some of the users will always be more knowledgeable than others, or have information others lack, so they can help out, no matter how fiery the environment. The line to the developers runs through the email lists and Trac (if you have an account). |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
But Miep is right to raise the question: why do they spend time tarting up the forums for 'us' to use, while 'they' commune elsewhere in some ivory tower? Aren't we all in this together? I know, that's the official line. But I honestly believe that the developers are being stupid by adopting that approach: they are wasting a valuable and under-used resource, namely volunteers' brain-cycles. You know that I use the mailing lists (on suffrance) as the only effective tool: I have a trac account (mainly only useful as a permanent record of how long ago I first reported an unfixed bug - I recently found myself reviewing [trac]#525[/trac]): and you know that I posted a long ramble on the subject of teamwork at the weekend - you sent me a PM about it, because you weren't posting on the boards at the time. And that PM was the *only* response I've received. It's a vicious spiral. The disdainful, patronising attitude shown by the developers and administrators to these boards discourages positive, engaged posters from trying to help: the negative emotions gain sway: everybody's preconceptions get reinforced: and we throw the baby out with the bathwater. Compare with the attitude shown by the devs and admins at AQUA. There the feedback is positive: crunchers' requests and reports are acknowledged and taken seriously. When things go wrong, as they did on Tuesday, the speed and openness of response turneth away all wrath. @ Mark, No, I won't say this in a PM - I'll say it out in the open here. I've not been choosing to ignore you: I've actually been ignoring everyone. A combination of burnout - I wrote something like 200 technical posts in the first two weeks after the Fermi app and 'new credit code' were loaded, with no feedback at all from the powers-that-be (see above): an exhausting pig of a work contract; and some personal matters to work through. Judging by the way this post is rattling out, there may be some light at the end of the tunnel - but it's still not certain it isn't an oncoming train. You have your own demons to face, my friend, and I hope we see you back when they've gone. |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
The biggest problem I see and why I can understand why the Powers That Be don't read these forums is how fast a truly helpful suggestion gets buried by a storm of negative posts, mostly having nothing to do with the original post. They may start out trying to add their support to the OP but very quickly deteriorates to arguments and total BS. Even I, as just an interested volunteer get sick of reading them sometimes. PROUD MEMBER OF Team Starfire World BOINC |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13847 Credit: 208,696,464 RAC: 304 |
Sorting the Wheat from the Chaff has always been a difficult & laborious process. Grant Darwin NT |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
Sorting the Wheat from the Chaff has always been a difficult & laborious process. So? What's the best response to that (very real) problem? Recruit some knowledgeable volunteer Wheat/Chaff sorters, and give them the tools to undertake the task? Or throw everything - Wheat and Chaff alike - onto the rubbish tip? |
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
... As your hosts are running stock applications, they're very interesting as examples of the kind of estimate adjustments the servers make. Most active forum members here are running anonymous platform, for which that feature is disabled for now. The "Number of tasks completed" needs to be at least 10 before BOINC will try to adjust estimates, and for a CPU getting to that point will clearly take longer than most GPUs. That period at the beginning for each host when CPU estimates are unadjusted while GPU tasks have the adjustment is certain to be awkward. If the estimates for your AMD CPU tasks with 7 completed are even worse than those for the Intel CPU tasks with 27 completed, perhaps that's a positive sign. OTOH, the servers are maintaining an overall host average as well as the app version averages, the relative speed of CPU versus GPU tasks on a host may also affect how long it takes for estimates to stabilize as much as they can. Please keep us posted on future developments, BOINC is trying for a 'one size fits all' method. That usually really means the fit is poor but maybe adequate for all. Joe |
Keith White Send message Joined: 29 May 99 Posts: 392 Credit: 13,035,233 RAC: 22 |
Okay. I don't lurk here that often so I may have missed the scuttlebutt about this but what's going on here? 7/9/2010 7:42:17 PM SETI@home Sending scheduler request: To fetch work. I uploaded 6 work units but I'm only allowed to download two then I'm blocked? I see people mentioning a 5 work unit per CPU limit but isn't that going to starve the mega crunchers? Right now I have about three days worth of work queued when my preferences are asking for a week's worth. Is this new per CPU quota get kicked up as we get closer to the outage? Is this some measure to prevent orphan work units from systems that fetch 20 days worth then are never seen again or have some catastrophic hardware problem? Is this their way to keep the number of work units in progress under control? Not complaining about the concept, more of a mixture of curiosity and seeking a better understanding how things work now. "Life is just nature's way of keeping meat fresh." - The Doctor |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.