Message boards : Number crunching : 2 million a day
Author | Message |
---|---|
![]() ![]() Send message Joined: 25 Oct 05 Posts: 576 Credit: 4,693,680 RAC: 38 |
One month ago yesterday Rosetta broke the "1,000,000 credits per day" mark, having been at half that level a month earlier. I just looked, and yesterday Rosetta (in the middle of 'bad WU' problems) did 1,845,924 credits. Two million a day is on the horizon now folks... I'm betting 1/4/06 for the "first time", and 1/15/06 for the beginning of a sustained 2M-and-up level. Any takers? :-) (BTW, poor Predictor only did 225,820... a project "supposedly" 3x the size of Rosetta just a short while ago... seems they've had some communications problems between staff and participants...) ![]() |
![]() Send message Joined: 17 Sep 05 Posts: 815 Credit: 1,812,737 RAC: 0 |
Well, I did my part, I upped my percentages, and added Rosetta@Home back to the PowerMac G5 ... of course it is only at 25% resource share ... but that ought to help ... And, in April when I get Einstein@Home over SETI@Home for total credit I will be taking their share down and stressing RAH more ... Of course, Bill, if you would just send me $3K I could get the Quad and this would all happen faster ... :) |
Scribe![]() Send message Joined: 2 Nov 05 Posts: 284 Credit: 157,359 RAC: 0 |
Lots of ex-FaD guys started coming in....... ![]() ![]() |
![]() ![]() Send message Joined: 15 Dec 05 Posts: 761 Credit: 285,578 RAC: 0 |
Lots of ex-FaD guys started coming in....... and a few people who left CPDN when they made their wu even bigger and a few Seti-classic folks who had been resisting BOINC, once they finally had to get BOINC just a few of them also looked further afield as well as starting SETI BOINC. This has certainly been the biggest ever influx to BOINC and it only takes a fraction of a perent of them to make a big difference to a still-small project. Three boosts all in the same month - but I agree the FaDs were probably the biggest single effect poor Predictor ... seems they've had some communications problems between staff and participants It will be interesting to see how -- if at all -- the recent problems affect Rosetta's sign-up rate and if it generates any noticable 'outflux' of people (accounts that were crunching regularly and now stopped -- just how do we select for that in Boincstats anyone?) Some people, including me, will be retained and encouraged by the careful involvement of the project team but my guess is that others will have already left without telling us - some permanently and some just till the dust sttles a bit. But yes - 2Meg by Y2k6 lets go go go... |
![]() ![]() Send message Joined: 15 Dec 05 Posts: 761 Credit: 285,578 RAC: 0 |
O1/4/06 for the "first time" Nahhh, I bet it happens before April 1st. ;-) Seriously - everyone not just Bill - please try to remember to avoid numerical dates they are ambiguous in an international forum. Jan-1-2006 or 1-Jan-2006 can't be mistaken even if to some of the readers it looks odd. R~~ |
Ingleside Send message Joined: 25 Sep 05 Posts: 107 Credit: 1,514,472 RAC: 0 |
One month ago yesterday Rosetta broke the "1,000,000 credits per day" mark, having been at half that level a month earlier. I just looked, and yesterday Rosetta (in the middle of 'bad WU' problems) did 1,845,924 credits. Two million a day is on the horizon now folks... I'm betting 1/4/06 for the "first time", and 1/15/06 for the beginning of a sustained 2M-and-up level. Any takers? :-) Of course, since Rosetta@home is giving a boatload of credits to an user that has errored-out all his results, hitting whatever target isn't really a problem... :innosent-looking: As for what time Rosetta@home will really hit 2.000.000 Cobblestones/day, your guess is likely better than mine, but would still guess on sooner than April. :) As for Predictor@home, low stats is normal when a project is having server-outages... Anyway, more interesting whan Rosetta@home hitting 2M Cobblestones/day is what will happen when Folding@home/BOINC starts their public beta-test "soon"... |
![]() Send message Joined: 17 Sep 05 Posts: 815 Credit: 1,812,737 RAC: 0 |
BOINC Stats does not track leaving, but does track the active vs. registered. Drill down on a project and you can get the graphs. For example Hosts by OS or Users for Rosetta. The second graph down shows active. So, when the time comes you can see if it is trending down. As an example, though it is hard to see it well, Predictor@Home though it is still picking up people daily, has a down trend. I know that I will show up as a down tick when 20 more days pass ... So, it looks to me like maybe they are having trouble keeping people? Might communication be a problem? |
![]() ![]() Send message Joined: 15 Dec 05 Posts: 761 Credit: 285,578 RAC: 0 |
BOINC Stats does not track leaving, but does track the active vs. registered. Drill down on a project and you can get the graphs. but you can't tell if the inactive people are newcomers (and soething like 50% of all newcomers to all DC projects never submit more than one piece of work) or whether they are seasoned crunchers who were valuable to the project. What I'd like to be able to ask is SELECT COUNT(*) FROM blah blah WHERE (CREDIT_NOW - CREDIT_LAST_MONTH) < 0.1 * (CREDIT_LAST_MONTH - CREDIT_MONTH_BEFORE)By the way, Willy, if you see this post, this is not a complaint - no package can possibly do everything. River~~ ![]() |
Ingleside Send message Joined: 25 Sep 05 Posts: 107 Credit: 1,514,472 RAC: 0 |
BOINC Stats does not track leaving, but does track the active vs. registered. Drill down on a project and you can get the graphs. Well, seeing how #active users in LHC@Home has increased the last 5 days, then at the same time LHC@home haven't given out any credits for over 10 days, I won't expect the other graphs is more correct either... As for keeping users, looking on Toby's stats it seems all projects is losing users or only barely breaking even now during the holidays, so wouldn't draw any conclusions currently. How accurate #active are is another matter, since on this stats-site they're based on RAC and not on how many has really gotten any credits resently... |
![]() ![]() Send message Joined: 25 Oct 05 Posts: 576 Credit: 4,693,680 RAC: 38 |
Of course, since Rosetta@home is giving a boatload of credits to an user that has errored-out all his results, hitting whatever target isn't really a problem... :innosent-looking: The "zillion credit" guy is already excluded from both Willy's and Zain's sites... otherwise it'd be more like 900,000,000,000,000 credits for the day... And ok, I, who put USD$xxx referring to money in an attempt to not be "US-centric", _do_ try to make dates international-friendly, but sometimes slip. I would point out that there were two dates given in the same sentence; 1/4/06 and 1/15/06. If you want to make the first one April 1, that's fine... but what is the second one?!? ![]() |
![]() ![]() Send message Joined: 17 Sep 05 Posts: 161 Credit: 162,253 RAC: 0 |
I would point out that there were two dates given in the same sentence; 1/4/06 and 1/15/06. If you want to make the first one April 1, that's fine... but what is the second one?!? Probably something like ***UNHANDLED EXCEPTION**** Reason: Access Violation (0xc0000005) at address 0x00656358 read attempt to address 0x2A08CFE8 Sorry, couldn't resist :-) *** Join BOINC@Australia today *** |
Ingleside Send message Joined: 25 Sep 05 Posts: 107 Credit: 1,514,472 RAC: 0 |
The "zillion credit" guy is already excluded from both Willy's and Zain's sites... otherwise it'd be more like 900,000,000,000,000 credits for the day... Well... if 1/4/06 is 01.04.2006, to get things to go up it would mean 1/15/06 is 01.03.2007. :) Anyway, seeing production has dropped to 1,47M Cobblestones today, my guess on before April is likely still true, but not certain you'll meet your guess of 04.01.2006... |
![]() Send message Joined: 19 Sep 05 Posts: 403 Credit: 537,991 RAC: 0 |
One month ago yesterday Rosetta broke the "1,000,000 credits per day" mark, having been at half that level a month earlier. I just looked, and yesterday Rosetta (in the middle of 'bad WU' problems) did 1,845,924 credits. Two million a day is on the horizon now folks... I'm betting 1/4/06 for the "first time", and 1/15/06 for the beginning of a sustained 2M-and-up level. Well we passed the 2.000.000 credits now :) Server Status as of 21 Jan 2006 6:29:43 UTC [ Scheduler running ] Queued: 87,410 In progress: 269,628 Successes last 24h: 105,400 Users (last day ) : 36,535 (+214) Hosts (last day ) : 74,492 (+564) Credits last 24h : 2,064,877 Total credits : 109,007,851 TeraFLOPS estimate: 20.649 Anders n ![]() |
arcturus Send message Joined: 22 Sep 05 Posts: 16 Credit: 525,440 RAC: 0 |
Well we passed the 2.000.000 credits now :) Generous point awards + no validation = no surprise. Very alluring to the point counters who'll turn on a dime and whine if things change or things go awry. We saw a taste of that not long ago. |
![]() Send message Joined: 5 Nov 05 Posts: 98 Credit: 618,288 RAC: 0 |
yes its nice to see we broke a 2 mill a day @arcturus: yes people will find a way to abuse any system out there i remember back on seti (before clasic... before gold...) that one method people would cheat was right befor a job was complete (last save point) they would transfer it to 50 of there machines.. these 50 machines would complete from the last save point and upload and BANG lots of credits... the reason this project is non redundant is that as long as the data is sound then why cut your work output by 1/2 or 1/3 or... ![]() ![]() |
Message boards :
Number crunching :
2 million a day
©2024 University of Washington
https://www.bakerlab.org