81) Message boards : Number crunching : BOINC WU priority for projects (Message 77183) Posted 6 Feb 2005 by UBT - Timbo Post: > Why would the machines on dialup need a larger cache? Because they are slower PC's and I only get to see them about once a week (they are some distance from me). > Now, what I'd do is get one of the small DSL "routers" that has a modem-backup > feature. If you never connect the broadband connection, then it'll be > dialup-only, and your machines can easily share the dialup connection. I'll look into that, but as the PC's are remote to where the ADSL facility is, then I'm not sure what you propose would be any better than what I currently use (a single dial-up PC, a small hub and FreeProxy software to link the PC's together when they want to up/download). regards, Timbo

82) Message boards : Number crunching : BOINC WU priority for projects (Message 77107) Posted 6 Feb 2005 by UBT - Timbo Post: > This is what he meant by ”the number you set to connect to the server” > This number determine how many WU’s that are cached on your computer. If you > set it to 3, you will get 3 days of work stored on your computer. Now, BOINC > doesn’t take in consideration your resource share for each project, so you > will get 3 days of work stored fore each project. With 4 active projects you > will get 12(in reality 6-12) days of work stored on your computer. Thanks for that - trouble is this: 1 PC (the one in question) has the ADSL connection. All the others (which are a bit slower) are on dial-up. And so these need a larger cache to be stored locally (they upload/download about once a week - take about 2 hours to send the completed work and to download new WU's). So, the end result is a fast PC working on 4 current projects, but which BOINC is not allowing "soon to pass" deadline WU's to be crunched first. And a small farm of PC's crunching for 2 projects each, which need to be fed just once a week (in a separate location - otherwise they'd all be on ADSL !!). Guess I need to look again at how many projects each PC can best serve.... (or wait until BOINC get's it's act together and the home/work/school preferences start working correctly - are they still u/s?? - what I need to do is to set a different connection criteria for teh PC's on dial-up compared to the one on ADSL). regards, Timbo

83) Message boards : Number crunching : Rom Walton, User of the Day! (Message 77070) Posted 6 Feb 2005 by UBT - Timbo Post: > Congratulations Rom! :-)))) > > [url=http://setiweb.ssl.berkeley.edu/profile_menu.php]User of the Day > -- February 6, 2005<a> > Never has a User of the Day been more worthy. Well done ROM - fame at last !! regards, Timbo

84) Message boards : Number crunching : BOINC WU priority for projects (Message 77069) Posted 6 Feb 2005 by UBT - Timbo Post: > 2 factors are paying a role: > > 1. the resource share, > 2. the number you set to connect to the server. > > If your settings are done in an appropriate way, you should be able to get all > the WU's of any project to return the results on time. > > If you have some WU's past their dealine, either you downloaded too much WU's > either you should change your resource share. > > Hi Thierry, Tks for the reply. 1) Resource share: SETI, Predictor, LHC and Einstein all set at 22.5% (but with LHC non-operational, that means these should all be around 30% each approx) CPDN (having a very long deadline) set to 10%. Therefore, my assumption being that the WU time is around the same for these 3 projects, "should" mean that BOINC deals with these in deadline order. But what has happened is that I've already run out of Predictor units, BOINC is now switching between CPDN and SETI, leaving the Einstein WU's untouched (not 1 even part crunched). 2) the number you set to connect to the server I don't call any number, as I have an 1Mb ADSL (always on) net connection. 3) This might help - my BOINC network connection is set to connect every 1-3 days (but usually returns completed WU's straight away). 4) And we're talking about a 3GHz Pentium 4 here, not some slow 20Mhz 386 (if anyone remembers THEM !!) Does that help explain? regards, Timbo

85) Message boards : Number crunching : BOINC WU priority for projects (Message 77040) Posted 6 Feb 2005 by UBT - Timbo Post: Hi all, Wonder if some-one can advise me about how BOINC prioritises project WU's. I run BOINC v4.19 on a few PC's but on one, BOINC has prioritised (of it's own accord, apparently) that it should crunch certain projects WU's (CPDN, SETI, Predictor) where the "Report Deadline" is much later than the deadline for another project (Einstein). The result is that I've now missed the deadline to return a number of WU's for Einstein as a result of BOINC "deciding" that WU's with later deadlines should be crunched first. Surely, there should either be: a) a mechanism within BOINC to recognise when deadlines are approaching and to crunch those WU's first, or b) for the user to be given manual control of which WU's should receive priority. rgds, Timbo

86) Message boards : Number crunching : Alpha Client Release---POLL (Message 74358) Posted 26 Jan 2005 by UBT - Timbo Post: > It has been suggested that there are users of this message board that do not > understand that the Alpha released clients (IE 4.14, 4.15, 4.16, 4.17, 4.18, > and the like) possibly contain bugs, and that these bugs can do "Bad" things > to your computer and/or WU's. I therefore request that these Users who do not > know of the hazards involved PLEASE Post a note here. > Tony: One of the problems is that BOINC has always been about have multiple projects so you can be crunching for one when no WU's are available for another. And the biggest recent problems are that some projects have seemed to "break" BOINC and hence BOINC software has had to be fixed in order to resolve the issues. In my case, as I crunch on numerous machines for all 6 public projects, it then means that I am having to update each host to ensure that the fix is in place for the projects that need the fix (otherwise maybe the work done is useless??). So, if a problem is found with CPDN, then I have to update the BOINC client on machines that run CPDN (but they may also run other projects). Then if the need for a BOINC update is made by Einstein, then a further update has to be done on these machines. And if we get multiple new releases every week, then it becomes frustrating to try and keep everything uptodate. (It would make so much more sense for BOINC to auto-update - at last on those platforms that support that function - if anti-virus programs (and otehr s/w can do it, why not BOINC ??). And you are right to think that these projects need a central "system" to make crunchers aware of the need to update their s/w (assuming auto-updates aren't possible). After all, we all signed up with particular email addresses - so the data is there to tell people to update. rgds, Timbo

87) Message boards : Number crunching : OT: CPDN makes BBC Website top news story. (Message 74357) Posted 26 Jan 2005 by UBT - Timbo Post: Hi, The news that CO2 levels are likely to rise alarmingly high very soon has been published by the BBC news website. Seems us CPDN crunchers are famous now. Don't know if it'll mean that the yanks will at last take note and sign up for Kyoto. Timbo (copied to SETI BB as CPDN has a security problem with phpBB software issue).

88) Message boards : Number crunching : Win98se and Boinc time accumulation, when paused? (Message 74115) Posted 25 Jan 2005 by UBT - Timbo Post: > I couldn't find anything at Einstein except for > Sign up for email. > No forums or anything. > > Have a link please? > > click here Timbo (PS Forgot to mention - membership is currently by invitation only as it's currently in TEST only - should be live sometime soon tho').

89) Message boards : Number crunching : Win98se and Boinc time accumulation, when paused? (Message 74101) Posted 25 Jan 2005 by UBT - Timbo Post: > I have Climate and Seti running on Boinc. > If one is running the other is paused, no brainer, but > I often catch both the running and paused cpu times going up at the same > time. > > Anyone else see this? > Have one Seti result at 57,xxx seconds. 2800+ cpu. > Hi, This is a known bug. I've reported it and so have others. It was thought to be a Win9x specific bug, but it seems that the dev's at Einstein@home, have produced a client "exe" that when the project is paused on Win9x systems, the CPU time is also paused. At present, all other projects do NOT pause the CPU time, when the project is paused, under Win9x. As such, all of us still running a Win9x box are seeing our "total credits claimed" for each WU, be substantially higher than the same type WU if crunched on a Win NT/2k/XP box. I don't think many people are running Win9x as the dev's do not appear to have made this a high priority fix. Maybe it will get fixed. (Surely one of the stats websites can come up with some numbers/percentages of all the BOINC hosts as to how many are running Win9x ????). Timbo

90) Message boards : Number crunching : server probs?? (Message 73094) Posted 22 Jan 2005 by UBT - Timbo Post: Am trying to access my SETI@BOINC account page but only see the following in the browser window: can't read ../cache/12/team_display.php_teamid%3D112288%26sort_by%3Dexpavg_credit%26offset%3D0; lastmod 1106430445 Is this the beginning of a server problem? rgds, Timbo

91) Message boards : Number crunching : Win 98 cpu timer bug isn't a Boinc problem, it's an App problem (Message 72186) Posted 20 Jan 2005 by UBT - Timbo Post: > On my win98 machine I'm running 3 apps. Seti,Ppah, and Einstein. The "Work" > tab shows all three. since they are sharing time, one project runs while the > other two are paused. When LHC, Seti, and Ppah are paused the "CPU Time" > continues to increase, however this doesn't happen when the Einstein > computation is paused. The Einstein app seems to be operating the way the > others should. It operates the way my XP machines do. > > Could there be a solution to the problem coming now that it's narrowed down? > Maybe its not important enough since this old machine is close to death. lol > Hi, I've been succcessfully running BOINC (all versions from v3.xx upto 4.15) and on three/four projects (SETI, LHC, CPDN, ProPredictor) since back in Sept 2004 on a 3.06GHZ Pentium 4 running Win98SE. Until Einstein invited me to join the project a couple of days ago, the time shown as "CPU Time" (within BOINC) for "paused" projects was always increasing, but only AFTER the first time each project had run after the BOINC program had started. Or in other words, until (say) SETI had been started, it's CPU time was fixed. But once BOINC had re-started the project (say after a re-boot) then the SETI "CPU time" would only carry on increasing after the SETI program had been paused by BOINC (and of course BOINC has switched to a different project). Hence over three/four projects the CPU time would be increasing on all projects once BOINC had allowed each of them to run. (Hope some one can understand the above - if not pls reply to this msg). BUT.... With Einstein (v4.71), even after the project has run the first time, it's CPU time is paused and only carries on once the projects is re-activated by BOINC. So, methinks that the Einstein team may have fixed the so-called "Win98 bug" which causes said clients to claim far more credits than they are actually due. regards, Timbo

92) Message boards : Number crunching : Website main page (Message 71669) Posted 19 Jan 2005 by UBT - Timbo Post: > Not to steal any of Janus' deserved thunder (he's done A LOT for Boinc, most > recently the moderation code)... but it wasn't Janus who redesigned the site. > > If you look at the bottom of the front page, you'll see it was designed by <a> href="http://effect-web-media.com/">Effect-web-media.com[/url]... apparently > they donated the site design. :-) > Just hope the main sponsors aren't "peeved" that their logo's are not on the front page anymore and instead have a seperate link page called "sponsors". Timbo

93) Message boards : Number crunching : Check your email everyone! (Message 71418) Posted 18 Jan 2005 by UBT - Timbo Post: > I just got an email from Einstein@home asking me to help test the project! > Maybe you other people who, like me, signed up but didn't get a test slot got > one now! > > :D :D :D > I'm in - thanks for the info - been waiting to sign up since it was first announced (many moons ago) Timbo UK BOINC Team

94) Message boards : Number crunching : Log files (Message 70812) Posted 17 Jan 2005 by UBT - Timbo Post: > I am working on error messages in the new FAQ database and I have gone through > all my on-hand logs and have captured all of the messsages that I can find. > > I know I am missing some. SO, I am asking if anyone has log files with the > messages in them that I can scan. > > Alternatively, you can look in the database to see which ones I have already > ... > > If you have log files, I sure could use them now. > > Thanks! > I had this when ProPredictor failed this morning: --- - 2005-01-17 09:53:15 - FILE_INFO::~FILE_INFO(): removing FILE_INFO when a pers_file_xfer still points to me HTH, Timbo

95) Message boards : Number crunching : Protein Predictor @ home Status message (Message 70811) Posted 17 Jan 2005 by UBT - Timbo Post: > > Currently, I get constant download errors. In order to avoid a mess, I > > detached for the moment being. > > ditto: BIG SNIP > ProteinPredictorAtHome - 2005-01-17 20:58:36 - Unrecoverable error for result > t0201D_1_7893_2 (WU download error: couldn't get input files: > t0201D_1_7893.seq: MD5 computation error I also had this problem and to compound it all, it locked up BOINC such that I couldn't d/l any NEW WU's from SETI and CPDN. And their website went down, so nothing could be accessed, nor comments made to their webmaster about the problem. SO, like the OP, I too had to detach just to clear my PC. God knows how many WU's the scheduler has tried to dl/ to clients machines today - and mostly (I guess) they've all failed..! What a waste! Timbo

96) Message boards : Number crunching : Speed up seti (Message 70760) Posted 17 Jan 2005 by UBT - Timbo Post: > How on earth can I speed up my setiBOINC on my Win98SE computer? Anyone? > 1) Get a faster computer or if not possible: 2) Get a faster CPU 3) Get some more (faster) memory 4) Change OS as Win98SE gives the wrong "processing time" for each WU it's processes (this will lead to higher credits being given) rgds, Timbo PS Almost forgot: Join our team !!

97) Message boards : Number crunching : Sunday. (Message 69797) Posted 15 Jan 2005 by UBT - Timbo Post: > I would agree, the best way to handle trolls is to ignore them. > But these new guys are like the EverReady bunny.... > they keep going, and going, and going.... > Hi DogBytes Not trying to be pedantic, but wasn't it the Duracell bunny....? (Credit where credits due, after all ;) ) Timbo

98) Message boards : Number crunching : BOINC client for Mac Mini? (Message 63325) Posted 12 Jan 2005 by UBT - Timbo Post: Just heard about the new Mac Mini, which will sell for UK GBP 339 / US $ 499 (but without display/mouse/keyboard). The basic Mac Mini uses a G4 running at 1.25Ghz/1.42Ghz (depending on which one you buy) uses an ATi 9200 graphics engine (w/32Mb video RAM) and the OS is Mac OS X 10.3 (Apple sales pitch to off). So I'm guessing that BOINC should run on it ! Any one care to throw some light on this (EDIT: ie the principle of whether BOINC would run on it? /EDIT Timbo

99) Message boards : Number crunching : OT: Predictor site now open to create new accounts (Message 62880) Posted 10 Jan 2005 by UBT - Timbo Post: The Predictor site now seems to allow crunchers to create new accounts: <a>http://predictor.scripps.edu/create_account_form.php[/url] or <a>http://predictor1.scripps.edu/create_account_form.php[/url] Tim

100) Message boards : Number crunching : Highest # of claimed credit (Message 62292) Posted 8 Jan 2005 by UBT - Timbo Post: > It's called Free Enterprise. IF you were American and IF you created > something other than a complaint, then you could sell it for whatever price > you wanted as well. In order to maximize your profit margin, you want the > price as high as your consumers are willing to pay. > > If your product was wildly successful, like windows, then you could sell it > overseas, but some countries, like the UK and the USA for instance, impose > tarriffs on all imported goods. > > In other words, it's YOUR countries fault that you have to pay so much for > windows. It's the EXACT same reason a Mini Cooper cost nearly $30,000 in the > USA. Well a Mini Copper in UK sells for around GBP 16,000-18,000 depending on spec. At current exchange rates that's around $30,000...! US buyers aren't getting such a bad deal when you think it, as that $30,000 includes shipping the damn things across the pond. Compare that to a US product, where most US makers don't even understand what an export distributor is...! End result? US products sell in UK/Europe for the same figure but with the $ change to STERLING. And as for MS having such a great product that market forces/consumer demand would help keep it's price up? I can't see that. More like the UK distributor (a MS-owned subsidiary company) wants to screw more out of non-US buyers....! Tim

Previous 20 · Next 20

 
©2020 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.