Message boards :
Number crunching :
When is Seti classic going away
Message board moderation
Author | Message |
---|---|
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
I have been running Seti classic for close to four years now and have with the large batch of computers configured to run the command line interface generated a total of over 140,000 Seti classic units. I've got a two parter question. 1) When will SETI classic actually be shut down (I've heard dates ranging from 12/2004 (manifestly not the case) to sometime in March or April of this year (but suspect it might actually be later than that). 2) When that day happens -- will all of the classic units be converted for 'credit' in BOINC world. I have set up a BOINC account and have watched the ongoing tribulations of an extended beta test being run with a very large and diverse user population. So my sense is that things won't 'settle down' for some time yet and that SETI classic might stretch out to the late Spring -- (giving me time to get to the top 500 of all Seti Classic users). But that may be just wishful thinking on my part. Barry Schnur |
B-Roy Send message Joined: 4 May 03 Posts: 220 Credit: 260,955 RAC: 1 |
> > 2) When that day happens -- will all of the classic units be converted for > 'credit' in BOINC world. > Certainly not, because the credit system at boinc is completely different to the old one. your amount of wus done under classic is however displayed in your account. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13755 Credit: 208,696,464 RAC: 304 |
> 1) When will SETI classic actually be shut down. When the new BOINC version is ready to handle the load. Grant Darwin NT |
Jaaku Send message Joined: 29 Oct 02 Posts: 494 Credit: 346,224 RAC: 0 |
> 2) When that day happens -- will all of the classic units be converted for > 'credit' in BOINC world. As B-roy said no they won't. The credits you had on the 14th of may will be in listed in your area. Like mine "SETI@home classic workunits as of 14 May 2004 401 SETI@home classic CPU time as of 14 May 2004 2,085 hours" But some people have said that they may update it when classic is closed! |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
|
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
> As B-roy said no they won't. The credits you had on the 14th of may will be in > listed in your area. Like mine > > "SETI@home classic workunits > as of 14 May 2004 401 > SETI@home classic CPU time > as of 14 May 2004 2,085 hours" OK -- thanks for the detail -- so in terms of 'credit' Seti Classic and Seti BOINC are two different ventures. I suppose in terms of other statistics, the BOINC numbers will also be a start from 0 sort of thing -- that is, that the BOINC seti project itself won't list a membership count of 'classic only' folks at the individual, group, country or total statistics level. > > But some people have said that they may update it when classic is closed! > I recall back in the fall that the early BOINC advocates seemed pretty clear in saying that the classic unit count would be updated and brought forward into the BOINC project -- I suspect in dealing with all the other complexities encountered during the ongoing beta, some functionality needed to be trimmed. Thanks for the answer though. Barry Schnur |
Matt Lebofsky Send message Joined: 1 Mar 99 Posts: 1444 Credit: 957,058 RAC: 0 |
There are several ducks we need to get in a row, but basically any day now we'll embark on the first step, which is to tell all new users to sign up for BOINC instead of the classic SETI@home. Some of the aforementioned ducks include another database server transfer, a whole server closet reorganization, some web site updating, last minute bug fixing (and an ensuing new client release), a mass e-mail to all the old SETI@home users, etc. This can (and is) actually all happening in parallel. The main objective (a new database server) has already been completed and it is handling 25% of the expected full load without breaking a sweat. - Matt -- BOINC/SETI@home network/web/science/development person -- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
> As B-roy said no they won't. The credits you had on the 14th of may will be in > listed in your area. Like mine > > "SETI@home classic workunits > as of 14 May 2004 401 > SETI@home classic CPU time > as of 14 May 2004 2,085 hours" |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
> There are several ducks we need to get in a row, but basically any day now > we'll embark on the first step, which is to tell all new users to sign up for > BOINC instead of the classic SETI@home. > OK -- thanks for that. I did sign up for BOINC in the fall, but decided to remain in the classic world until those aforementioned ducks were well and truly lined up. I do realize the chore that running in parallel for an extended time can be (in my youth I did user project management of some fairly complex mainframe software implementations) > Some of the aforementioned ducks include another database server transfer, a > whole server closet reorganization, some web site updating, last minute bug > fixing (and an ensuing new client release), a mass e-mail to all the old > SETI@home users, etc. > This can (and is) actually all happening in parallel. And thanks for that detail -- I sort of wonder what sort of 'yield' BOINC Seti will get out of the classic folks -- realizing that of the 5.5 million 'users' listed in Seti Classic a very large proportion have long since 'retired'. > Thanks for the detailed reply. Regards, Bary Schnur |
Borgholio Send message Joined: 2 Aug 99 Posts: 654 Credit: 18,623,738 RAC: 45 |
> There are several ducks we need to get in a row, but basically any day now > we'll embark on the first step, which is to tell all new users to sign up for > BOINC instead of the classic SETI@home. > > Some of the aforementioned ducks include another database server transfer, a > whole server closet reorganization, some web site updating, last minute bug > fixing (and an ensuing new client release), a mass e-mail to all the old > SETI@home users, etc. > This can (and is) actually all happening in parallel. > > The main objective (a new database server) has already been completed and it > is handling 25% of the expected full load without breaking a sweat. > > - Matt > But the question is, can it handle 100% of the expected full load? :) As I understand it, the number of BOINC users is less than 25% of the number of Classic users. Or am I full of it again? You will be assimilated...bunghole! |
Astro Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
> > The main objective (a new database server) has already been completed and > it > > is handling 25% of the expected full load without breaking a sweat. > > > > - Matt > > > > But the question is, can it handle 100% of the expected full load? :) As I > understand it, the number of BOINC users is less than 25% of the number of > Classic users. Or am I full of it again? > I read Matts' comments this way. The current Boinc users are tasking the new server to 25% of its' capacity. That means that Boinc can take on 3 times the current number of users more than it has(whatever that is). any other thoughts? tony |
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
> I read Matts' comments this way. The current Boinc users are tasking the new > server to 25% of its' capacity. That means that Boinc can take on 3 times the > current number of users more than it has(whatever that is). > according to setisynergy.com the total numbers of Boinc users is 72,589 as of 19 hours ago. |
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
> As B-roy said no they won't. The credits you had on the 14th of may will be in > listed in your area. Like mine > > "SETI@home classic workunits > as of 14 May 2004 401 > SETI@home classic CPU time > as of 14 May 2004 2,085 hours" > > But some people have said that they may update it when classic is closed! > Rom Walton has said several times that the numbers will be updated AFTER Classic is closed and they are finalized. |
Astro Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
> according to setisynergy.com the total numbers of Boinc users is 72,589 as of > 19 hours ago. > So, if my guess is right, then 4x72,589=290,356 users is 100% load. That means the server can accept up to 217,767 new/classic transferred users. |
Saenger Send message Joined: 3 Apr 99 Posts: 2452 Credit: 33,281 RAC: 0 |
> according to setisynergy.com the total numbers of Boinc users is 72,589 as of > 19 hours ago. according to S@H classic about 420000 active users. That's about 6 times the amount of Boincys. Gruesse vom Saenger For questions about Boinc look in the BOINC-Wiki |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
> > > is handling 25% of the expected full load without breaking a sweat. > I read Matts' comments this way. The current Boinc users are tasking the new > server to 25% of its' capacity. That means that Boinc can take on 3 times the > current number of users more than it has(whatever that is). "Without breaking a sweat" means without effort -- it means it is handling it easily, with lots room to spare. It seems to follow that a fourfold increase would be handled easily. |
Speedy67 & Friends Send message Joined: 14 Jul 99 Posts: 335 Credit: 1,178,138 RAC: 0 |
> So, if my guess is right, then 4x72,589=290,356 users is 100% load. That > means the server can accept up to 217,767 new/classic transferred users. I think with for example only 100 users 'BOINCing' the database server still would have a load well over 0%, say 15%, with all the processes running. An increase of 72,000 users would therefore not automaticly mean a 25% increase of load on the server, maybe only a 10% increase. But then again, i'm no expert, so i could just as well be wrong. :) Edited: sorry, missed Matt's original post about the 25%. He's stating that 25% of the expected full load is being processed without the new server breaking a sweat. Which i guess doesn't mean the server itself is using 25% of its total performance. |
Matt Lebofsky Send message Joined: 1 Mar 99 Posts: 1444 Credit: 957,058 RAC: 0 |
Allow me to clarify: When originally planning load on our system, we determined that we needed to at least be able to send out one million results a day. This was our goal. Right now we are generating and sending out 250,000 results a day without taxing our database server. Will there be other unexpected bottlenecks? Probably. We'll find and fix those as new users ramp up. What happens if user interest is so high that there is a greater demand than one million results a day? Maybe we'll generate more, but remember the real beauty of BOINC (and one of the reasons it was created) - you can crunch on other projects if any particular project is temporarily out of work. - Matt -- BOINC/SETI@home network/web/science/development person -- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude |
Astro Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
So, basically they are planning on 4x the current users. If we get to 1 million wu/day, then we've satisfied the wants of project. it's kinda cool thinking that Boinc could be classified a success. if success is the users meeting the expectations of the project. cool |
Borgholio Send message Joined: 2 Aug 99 Posts: 654 Credit: 18,623,738 RAC: 45 |
> Allow me to clarify: > > When originally planning load on our system, we determined that we needed to > at least be able to send out one million results a day. This was our goal. > Right now we are generating and sending out 250,000 results a day without > taxing our database server. > > Will there be other unexpected bottlenecks? Probably. We'll find and fix those > as new users ramp up. > > What happens if user interest is so high that there is a greater demand than > one million results a day? Maybe we'll generate more, but remember the real > beauty of BOINC (and one of the reasons it was created) - you can crunch on > other projects if any particular project is temporarily out of work. > > - Matt > Ah I see! So your not basing server performance on the number of users, but rather on the amount of work you send out. That clears things up quite a bit! You will be assimilated...bunghole! |
©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.