Message boards :
Number crunching :
9 days
Message board moderation
Author | Message |
---|---|
Ray_GTI-R Send message Joined: 17 May 99 Posts: 56 Credit: 276,906 RAC: 0 |
The last time any of my farm contacted this project was nearly 9 days ago 20 Sep 2010 23:42:52 UTC Ultimately it will all work again as it always has. (Thread Posted just in case other noise hereabouts drowns out the Big Picture! FWIW I have no axe to grind, no records to break and humble admiration for the fact that in order to upgrade hardware a charity drive was recently, sucessfully completed FWIW2 The main Project sponsors are ... http://setiathome.berkeley.edu/sponsors.php FWIW3 I wonder how many 000 old/new project participants lost interest in the past 9 days. No reply required.) The difference between 0 and 1 is greater than the difference between 1 and 1,000,000 |
Ian Green Send message Joined: 25 Jul 10 Posts: 24 Credit: 102,337 RAC: 0 |
I guess the staff are distracted with other issues to get around to fixing the servers. That or they need some new servers. or the software they have is dodgy. |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
I guess the staff are distracted with other issues to get around to fixing the servers. This is some major admin work. I am sure they made progress. As far as how many have retreated.. It does not look like it is critical numbers yet. Although the disgruntled list has certainly grown. http://boincstats.com/stats/project_graph.php?pr=sah&view=users Janice |
W5DMG - Dave Send message Joined: 19 May 99 Posts: 155 Credit: 33,162,251 RAC: 0 |
I have several hundred units processed that have a report deadline of September 30th. Hopefully I will still get credit for them. |
Wiggo Send message Joined: 24 Jan 00 Posts: 34744 Credit: 261,360,520 RAC: 489 |
I've still got a few days worth left out of my 10 cache on my workers here, then once each machine has reported its results I'll detach/reattach it and clear a few ghosts out of the system. ;) Cheers. |
razamatraz Send message Joined: 23 Oct 07 Posts: 142 Credit: 27,815,748 RAC: 0 |
yup 1600 ready to report on the one machine and maybe half a day left in the cache....at least I'll have a chance to clean it out, this machine needs an OS rebuild anyways. |
Bearcat Send message Joined: 10 Sep 99 Posts: 106 Credit: 10,778,506 RAC: 0 |
Slowly but surely the completed wu's are going down the toilet. If I remember right units reported after deadline give no credits and will be discarded, right ? I am not worried about the credit, but I am not going to leave the comps running just to have the completed work discarded because the project is unable to receive it. It looks like I am having only a handful so far, but if no upload is possible by Oct 1, there will be a lot more than just a handful of units. |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
If I remember right units reported after deadline give no credits and will be discarded, right ? Not entirely true. In order for you to not receive credit for a valid workunit; a different copy must be sent to another computer, crunched and returned before you return your result. Since no one can upload, this would make it impossible for someone to upload before you. The moment the upload gates are opened, the reporting of work will be resumed and everyone will get their due credit. |
Bearcat Send message Joined: 10 Sep 99 Posts: 106 Credit: 10,778,506 RAC: 0 |
If I remember right units reported after deadline give no credits and will be discarded, right ? well, that sounds better. Thanks for the info. I never really paid much attention to the details on how such things work. |
52 Aces Send message Joined: 7 Jan 02 Posts: 497 Credit: 14,261,068 RAC: 67 |
well, that sounds better Assuming the converse that you'll be potentially issued re-issued WU's that will none-the-less be reported in ahead of you, but you'll go ahead and crunch them anyhow. Not a problem, per se, as I beleive the 3rd & 4th wingmen still get credit, but it doesn't advance the overall crunch picture as quickly as it might otherwise do. |
ScarabDrowner Send message Joined: 13 Sep 03 Posts: 90 Credit: 456,378 RAC: 0 |
http://boincstats.com/stats/project_graph.php?pr=sah&view=users This is depressing: Users: Total:1,126,281 Active:158,335 |
Aurora Borealis Send message Joined: 14 Jan 01 Posts: 3075 Credit: 5,631,463 RAC: 0 |
http://boincstats.com/stats/project_graph.php?pr=sah&view=users Not really... the Total represents the number of Boinc/Seti account ever created since the start. Many are duplicate accounts, lost passwords or other problems. About 10% of accounts have returned little or no work and were only attached for a few days if at all. Boinc/Seti active membership peeked at around 300,000 several years ago when there were only a couple of other projects. People come and go. We get new people and returning crunchers all the time. Some just move on to other projects with more short term gratification or more immediate goals. Seti is a major conduit to other projects membership, as would be expected. The current active users have much more powerful machine and do a lot more work. Despite AstroPulse and the search sensitivity being increased by a least a factor of 10, we are processing and continuously returning an ever increasing number of results. Boinc V7.2.42 Win7 i5 3.33G 4GB, GTX470 |
Bill Walker Send message Joined: 4 Sep 99 Posts: 3868 Credit: 2,697,267 RAC: 0 |
http://boincstats.com/stats/project_graph.php?pr=sah&view=users I think the total number of users is only part of the big picture. Just think about how fast the typical computer was 5 years ago, and how fast it is today. The total science being crunched is still all S@H can generate. The main topic on these threads today seems to be "why can't I get more work?". Seems to me that getting more users (or even retaining the existing users) won't solve that problem. |
Fred J. Verster Send message Joined: 21 Apr 04 Posts: 3252 Credit: 31,903,643 RAC: 0 |
But it's the number of hosts, rather then the number of users, banging on the SETI UP-- & D'Load SERVERS Users 1,126,281 158,335 active Hosts 2,731,841 245,695 active Teams 58,217 14,795 active Countries 234 206 Total Credit 83,123,681,683 Recent average credit RAC 65,321,486 Average floating point operations per second 326,607.4 GigaFLOPS / 326.607 TeraFLOPS. And the Methods used to perform certain tasks, like GPUGrid uses a <report_result_inmediatly>1</report_result_inmediatly> , to get a 'Bonus'. Those 245,695 hosts, of which probably 80% use/have (A)DSL or even Fiber, but maybe I'm totally wrong, in my estimate, are hammering constantly on the UP- and Down Load SERVER's, until, they get through and start a Down or Up Load, in the meantime, whithout intervening in BOINC's network function, it'll try to establish a connection....... A coïnsedense, which I don't believe, BOINC (6.10.56 on my Q6600+GTS250, running VISTA 32BIT, has UPLoaded all WU's, but no XP host(X64 & x86) didn't get through. (Probably changes or added network functions in VISTA?!) Since they are all on the same network-access-point (MODEM/ROUTER), all 32 bit hosts are running BOINC 6.10.56, only my XP64 (FERMI) is on BOINC 6.10.58.. Only download server 2 vader is Running. Do hope the crew will solve these issues, as soon as possible, yesterday morning, I ran out of SETI work and GPUGrid and Einstein, quickly filled the 'spare CPU & GPU cycles' work-qeue, which is a BOINC task! Maybe Distributed Storage and Splitting Data to Work Units and Completed Work Units Storage, is a next and not exactly an easy job......... |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
I guess the staff are distracted with other issues to get around to fixing the servers. They do indeed have a new server on the way made possible by a couple of recent donation drives. For more information, check the Technical News link on the home page, please. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Donald L. Johnson Send message Joined: 5 Aug 02 Posts: 8240 Credit: 14,654,533 RAC: 20 |
I guess the staff are distracted with other issues to get around to fixing the servers. Two things, in reverse order: 1) As Mark said, there IS a new, highly-capable server on the way, thanks to a group of S@H crunchers who were able to back-up their love for this project with some cold, hard cash. Once it comes on-line (by the end of October, we hope) several of the older, less-capable, less-stable servers will be retired. Until then, the S@H staff ARE working very hard to keep the current servers on-line in the face of an onslaught of state-of-the-art, multi-CPU/multi-GPU crunching rigs that are overwhelming both the current servers and the available data-transmission bandwidth. 2) The Seti@Home staff is not dedicated solely to S@H. All of the key people (Eric, Jeff, Matt, even Dr. Anderson) are employees of the UC Space Sciences Lab, and divide their time between S@H and other projects. They do spend a lot of their time keeping S@H running, including coming in on weekends and holidays to deal with casualties. Also, S@H is NOT a fully-funded University research project like most of the projects being run by the SSL. They get some grant funding, but most of their hardware and some of their admin costs are paid for by donations from those who participate in the project (us). They do an extremely good job given the resources available to them. Donald Infernal Optimist / Submariner, retired |
Ray_GTI-R Send message Joined: 17 May 99 Posts: 56 Credit: 276,906 RAC: 0 |
For more information, check the Technical News link on the home page, please. That's 5 days old news (with ultimate respect to any organiser of the fund drive, the SETI project et al). The great info provided by Donald L. Johnson should really be on the SETI@HOME front page, permanently. It may be common knowledge to old hands on this project ... but newbies need this up front in order to ''manage expectations''. And it would cut down on the 000's of DUH questions hereabouts recently and historically. Still and all ... >9 days now. The difference between 0 and 1 is greater than the difference between 1 and 1,000,000 |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
From Jeff Cobb in Tech news... Upload service is succesfully migrated. We now have over 4TB of space for uploads! Not that we'd ever use that much. There are still a couple of kinks to work out with other minor server migrations. Thus, we will leave the back end (validation, etc) down tonight. The splitters and data service are up. PROUD MEMBER OF Team Starfire World BOINC |
Ray_GTI-R Send message Joined: 17 May 99 Posts: 56 Credit: 276,906 RAC: 0 |
We now have over 4TB of space for uploads! Not that we'd ever use that much. Hands up who has heard a similar story before? I remember a friend buying a 170mb hard disk for about the same sort of price that 4TB now costs saying pretty much the same thing. Expectation counts for nought ... Anyway, it seems that servers are gradually (in the past few minutes) being brought back up. I got new tasks about 36 minutes ago. Pause for deafening applause! The difference between 0 and 1 is greater than the difference between 1 and 1,000,000 |
Blurf Send message Joined: 2 Sep 06 Posts: 8962 Credit: 12,678,685 RAC: 0 |
The Seti@Home staff is not dedicated solely to S@H. All of the key people (Eric, Jeff, Matt, even Dr. Anderson) are employees of the UC Space Sciences Lab Correction---Dr Anderson is actually paid for under a Boinc grant exclusively-he does not work directly for Seti. This was confirmed by Eric Korpela to me recently. |
©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.