Message boards :
Number crunching :
Well my 10 day cache is running dry...
Message board moderation
Author | Message |
---|---|
hiamps Send message Joined: 23 May 99 Posts: 4292 Credit: 72,971,319 RAC: 0 |
Too bad I didn't set for 10 + 10...Will keep the 10 day setting. My CPU still has a few but my room sure is getting cold. Am counting on Seti to keep my room warm this winter, trying not to turn the heat on in here all Winter. Hopefully things will get better soon. Official Abuser of Boinc Buttons... And no good credit hound! |
Salvis Cerins Send message Joined: 12 Dec 05 Posts: 5 Credit: 2,675,954 RAC: 0 |
|
Zeus Fab3r Send message Joined: 17 Jan 01 Posts: 649 Credit: 275,335,635 RAC: 597 |
I never heard that is possible to get WU for more then 10 days... I have 22 day cache of AP units on my FSC Lifebook. I left 3 days for safety measure cause AP's have 25 day deadline. If they increase that deadline, my cache will follow :-) Who the hell is General Failure and why is he reading my harddisk?¿ |
Uli Send message Joined: 6 Feb 00 Posts: 10923 Credit: 5,996,015 RAC: 1 |
I never heard that is possible to get WU for more then 10 days... You are right and welcome to the Forums. Pluto will always be a planet to me. Seti Ambassador Not to late to order an Anni Shirt |
PP Send message Joined: 3 Jul 99 Posts: 42 Credit: 10,012,664 RAC: 0 |
I never heard that is possible to get WU for more then 10 days... There are only 2 places in the code where the max limit is verified. You can of course change them and recompile the client. I haven't tried this particular modification myself so I can't say if there is a server limitation involved too. hiamps comment seems to imply it doesn't kick in at 20 days at least. |
Fred J. Verster Send message Joined: 21 Apr 04 Posts: 3252 Credit: 31,903,643 RAC: 0 |
Isn't BOINC handling this amount, whithout problems and whitout taking a lot of CPU-cycles? I remember, this was an issue, in previous BOINC versions! Well, I guess, I'll keep my 4 days cache, especially my VISTA host with an GTS250, makes some errors, so I gonna swap that card for a FERMI, too, when my 'finances' allow it. (First my promised Donation, but depending on one of my brothers, still having a job, is annoying enough.......) But in the meantime, keep on cruching, not your theets, though..... Reached 59 years today, why do they call it 'your bithday' since I'm already 58 years around ;-), feeling 'almost' old! Gonna visit my mother, she has Alzheimers and will become 90, in december. |
Uli Send message Joined: 6 Feb 00 Posts: 10923 Credit: 5,996,015 RAC: 1 |
Yes totally of topic Happy Birthday Fred. Pluto will always be a planet to me. Seti Ambassador Not to late to order an Anni Shirt |
Dave Send message Joined: 29 Mar 02 Posts: 778 Credit: 25,001,396 RAC: 0 |
+ all the best to your auntie. |
Fred J. Verster Send message Joined: 21 Apr 04 Posts: 3252 Credit: 31,903,643 RAC: 0 |
|
Julie Send message Joined: 28 Oct 09 Posts: 34053 Credit: 18,883,157 RAC: 18 |
|
hiamps Send message Joined: 23 May 99 Posts: 4292 Credit: 72,971,319 RAC: 0 |
Happy birthday Fred. Official Abuser of Boinc Buttons... And no good credit hound! |
BilBg Send message Joined: 27 May 07 Posts: 3720 Credit: 9,385,827 RAC: 0 |
I never heard that is possible to get WU for more then 10 days... Set "Connect ..." 10 days & "Additional ..." 10 days. (The first (Connect ...) can be more then 10 days) The "Additional ..." can also be set to more then 10 if you edit global_prefs_override.xml and edit the value to e.g.: <work_buf_additional_days>15.000000</work_buf_additional_days> but after that you can't close the Preferences window with [OK] (in case you want to change some other parameter) until you change the value (in the dialog) to 10 again. Â - ALF - "Find out what you don't do well ..... then don't do it!" :) Â |
Pappa Send message Joined: 9 Jan 00 Posts: 2562 Credit: 12,301,681 RAC: 0 |
I really do CAUTION AGAINST MONSTER Caches.... On any given day the forums are littered with bad advice. What follows next is tales of gloom and woe. I have been very quite of late, it time to say something simple. What is happening now: Common sense time... A Schedluer reply gets tromped on, a You have GHOSTS. With Ghosts you have drive space and Database bloat. With Ghosts Pendings go aboslutely nuts. With a Monster Cache, you have a machine that makes tons of "extra" requests for work that blocks others from reporting or causes the scheduler reply to be tromped on. With a Monster cache, you have a machine that can only report "part of what you have to report at a time (see extra requests from roughly 250000 computers). So as we think about all the Ghost Thread, We see Setizens complaining that the drives filled to max and Crashed! Now you are Stuck! Are helping to solve the issue or adding to it. Helping to solve the issue should be first (use common sense and be careful about the advice you listen to: otherwise I have a slightly used "Historical Bridge" for sale and promise that all proceeds will be donated to Seti. {transportation cost are extra}). My machines have not been able to report (2 of 4 one is Seti Beta; as everyone else is beating on the scheduler). I still refuse to take my cache above the 1 day cache. So while Seti will get there eventually, I have backup projects. IF you want advice, as we will be doing from Tuesday to Friday (doing science) work your cache to cover that period with a bit of room to spare. Then as the Server can actually catalog your resources you will get what you need. If a Server crash happens, it happens the staff will get things back running as quickly as possible. Then you are not adding some new unknown to the set of problems. Regards BTW: Happy "Day of Birth" Fred. The new yard blowers work great for blowing out that many candles (should you be caught short of breath) :) Please consider a Donation to the Seti Project. |
Ray_GTI-R Send message Joined: 17 May 99 Posts: 56 Credit: 276,906 RAC: 0 |
Mostly all OK now, from a me-perpective. Got new tasks, reported completed tasks and stats are updated. (As of about 40 minutes ago.) The difference between 0 and 1 is greater than the difference between 1 and 1,000,000 |
JohnDK Send message Joined: 28 May 00 Posts: 1222 Credit: 451,243,443 RAC: 1,127 |
I really do CAUTION AGAINST MONSTER Caches.... I for one will stay on a 5 day cache. I seem to remember someone saying that the max day cache could be set from the server, overriding the local preferences. How about setting it to max 5, 7, 10 or whatever staff seems is right. |
Zeus Fab3r Send message Joined: 17 Jan 01 Posts: 649 Credit: 275,335,635 RAC: 597 |
I for one will stay on a 5 day cache. They don't have to. There are other instruments for getting the job done. Several hours ago I was empty and now I've reached the limit of 26 cpu tasks! 01-Oct-10 04:00:56 SETI@home Requesting new tasks 01-Oct-10 04:00:56 SETI@home [sched_op_debug] CPU work request: 9339.04 seconds; 0 idle CPUs 01-Oct-10 04:00:56 SETI@home [sched_op_debug] CUDA work request: 0.00 seconds; 0 idle GPUs 01-Oct-10 04:02:22 SETI@home Scheduler request completed: got 0 new tasks 01-Oct-10 04:02:22 SETI@home Message from server: No work sent 01-Oct-10 04:02:22 SETI@home Message from server: This computer has reached a limit on tasks in progress Who the hell is General Failure and why is he reading my harddisk?¿ |
JohnDK Send message Joined: 28 May 00 Posts: 1222 Credit: 451,243,443 RAC: 1,127 |
They don't have to. There are other instruments for getting the job done. Several hours ago I was empty and now I've reached the limit of 26 cpu tasks! I don't think these limits is fair/good or whatever I should call it, it's the same limit whether you have a slow PC or a super fast one. |
Zeus Fab3r Send message Joined: 17 Jan 01 Posts: 649 Credit: 275,335,635 RAC: 597 |
They don't have to. There are other instruments for getting the job done. Several hours ago I was empty and now I've reached the limit of 26 cpu tasks! Nor is life but it still goes on :-) And I certainly understand why they have to apply them. Who the hell is General Failure and why is he reading my harddisk?¿ |
JohnDK Send message Joined: 28 May 00 Posts: 1222 Credit: 451,243,443 RAC: 1,127 |
They don't have to. There are other instruments for getting the job done. Several hours ago I was empty and now I've reached the limit of 26 cpu tasks! Yes after the outage, but not in general in my opinion. To take the max 26 CPUs as an example, on my netbook I guess it would mean 10 days worth or more, on my main PC I guess it's 2-3 days or something. |
Pappa Send message Joined: 9 Jan 00 Posts: 2562 Credit: 12,301,681 RAC: 0 |
They don't have to. There are other instruments for getting the job done. Several hours ago I was empty and now I've reached the limit of 26 cpu tasks! Thank You for bringing this up. The price of self protection is never fair! If you take time to look at the various stats sites you would know that around 248000 active computers are attacking Seti on a daily basis. If they are at the defaults it contact every 12 hours. That would mean they report work or request work. For simplicity I am going to round that to 250000 computers. 250000 x 2 = 500000 reports 250000 x 2 = 500000 requests 500000 + 500000 = 1000000 scheduler contacts per day. 1000000 / 86400 = 11.57 scheduler request / second Now it gets tougher, my current outbound sheduler request file is 303K (95 completed WU's). From the machine stats I know that its average upload rate is 34K/sec. When I can connect and report it will take a bit less than 10 seconds to complete. During that 10 seconds there will be 115.x computers connect attemting to upload or download fresh work or report. The router stats on previous outages have shown that just Scheduler requests can sustain 30 meagbits/sec (roughly 1/3 of the avialable bandwidth). Then we get to where a machine with 1000's of results to report is hammering away on the scheduler and trying to get more work... those 11.57 connections/sec is a very small number. No wonder it drops connections. This does not count as a machine completes a Workunit and starts an upload of a Result of roughly 30K. So without understanding all of the Client Server interactions that happen on a daily basis things do seem unfair. Regards Please consider a Donation to the Seti Project. |
©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.