Message boards :
Number crunching :
Hopefully...
Message board moderation
Author | Message |
---|---|
![]() ![]() Send message Joined: 4 Oct 07 Posts: 43 Credit: 53,898 RAC: 0 ![]() ![]() |
I've started 2 new searches, 3720282 and 3721282, in response to there being no new work; I don't quite understand yet, ho we ran out of work, but c'est la vie. I've increased the length of the WUs in an effort to ease the server traffic. I hope that this fixes a lot of the assimilator and traffic issues. Be sure to let me know if it works! As a heads up, I increased the time by increasing the accuracy with which we do the integral calculation over the wedge volume. It's possible that I made it too dramatic and increased the runtime by too much, so please let me know what completion times you are getting so I can adjust the next searches accordingly. I will also bring up the potential for needing a hardware upgrade to deal with the increase in traffic. I find this a good thing though as it means that we have all of you guys out there working with us. So I thank you for making this project such a success and I hope that we can get these few issues smoothed out for you as soon as possible. ~Nate~ |
![]() ![]() Send message Joined: 8 Apr 08 Posts: 45 Credit: 161,943,995 RAC: 0 ![]() ![]() |
looks like your new units run very long in comparison.. old units 151 sec for 3710182's new unit 3720282 1% completed after 178 sec 178s x 100 17800 nearly 5 hours even allowing for a fast last 50 % thats 2.5 hours that is an agressive change 2.5 hours for 4.06 credit be prepared for a loud whinge from all and sundry Good Luck Ross, Perth Australia Q9300 lUBUNTU linux commandline and headless OWN every thing I need EARN.. enough to live !!! WANT a solar array on the roof so I can run a BOINC farm( DREAM on!!) NO wife NO kids NO troubles |
![]() ![]() Send message Joined: 22 Dec 07 Posts: 13 Credit: 46,606,530 RAC: 0 ![]() ![]() |
The fastest I have completed so far is 10,612 seconds (just under 3 hours) on a stock Q9300 on 64-bit Linux. Previous tasks on this computer took about 160 seconds. Don't know what the credit is like as yet as so far everything has gone into pending. On my slower AMD X2 4200+ on 32-bit Windows XP, it is currently looking like taking 12 hours. Proud member of BOINC@AUSTRALIA |
![]() Send message Joined: 28 Aug 07 Posts: 146 Credit: 10,703,601 RAC: 0 ![]() ![]() |
I've increased the length of the WUs in an effort to ease the server traffic. I hope that this fixes a lot of the assimilator and traffic issues. Be sure to let me know if it works! As a heads up, I increased the time by increasing the accuracy with which we do the integral calculation over the wedge volume. It's possible that I made it too dramatic and increased the runtime by too much, so please let me know what completion times you are getting so I can adjust the next searches accordingly. Did you adjust the credits to that in the same amount, Nathan? Or are we getting the same 4 credits like everytime?!? Member of BOINC@Heidelberg and ATA! My BOINCstats |
![]() ![]() Send message Joined: 22 Mar 08 Posts: 65 Credit: 15,715,071 RAC: 0 ![]() ![]() |
hmm, ermm..mmkay LOL, I thought something was broke noticed MW units that had been running for a couple hours and was like...What the? o_O Ok, so got one finished with similar results as posted above: The old WU would take about 194sec clicky The new one took 11,658sec clicky Kinda funny they went from a hair over 3 minutes to a bit over 3 hours. Not sure about the credit yet it still says pending. Anyway, thanks for all the work you guys do. Best run project no doubt. Good luck with getting this sorted. Very curious about what happends with the credit. I have a feeling it may get a tad noisy on these boards over the coming days. Peace! -Brandon ![]() |
![]() Send message Joined: 28 Aug 07 Posts: 146 Credit: 10,703,601 RAC: 0 ![]() ![]() |
The new one took 11,658sec clicky Looking on that WU now says it got 260 credits! Wow, that's a pretty good outcome... I'm thrilled what my first of those will get (the hour average I mean ;-)) Member of BOINC@Heidelberg and ATA! My BOINCstats |
![]() ![]() Send message Joined: 22 Mar 08 Posts: 65 Credit: 15,715,071 RAC: 0 ![]() ![]() |
The new one took 11,658sec clicky WOOT! Too cool man. Thanks for pointing that out. Ok, I'm stoked. Everything's perfect guys, don't touch a thing. ;) ![]() |
![]() ![]() Send message Joined: 31 Aug 07 Posts: 66 Credit: 1,002,668 RAC: 0 ![]() ![]() |
It's possible that I made it too dramatic and increased the runtime by too much, so please let me know what completion times you are getting so I can adjust the next searches accordingly. I've got a new WU running on a slow P4. Usual completion time is around 11 minutes for that machine. Looks like the new WU is going to run for 21 hours! I'll let it run to completion and get a proper run time (assuming that it checkpoints nicely, if not I'll have to kill it). I hope the credit is more than the 260 quoted eleswhere in this thread, as that's going to be a bit mean compared to a lot of other projects ;) Al. |
![]() ![]() Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 ![]() ![]() |
Credits are fixed so you won't get more than 260 credits. Anyhow, i do like the longer crunch times cuz that should be quite a relieve for the poor server. @Nathan job well done! Thanks. ![]() Join Support science! Joinc Team BOINC United now! |
Send message Joined: 29 Aug 07 Posts: 486 Credit: 576,548,171 RAC: 0 ![]() ![]() |
Anyhow, i do like the longer crunch times cuz that should be quite a relieve for the poor server. I agree, I like short Wu's, but if the Server can't provide enough Work to the Participants or the Server can't take the strain being placed on it by the Participants then you may as well go to longer Wu's and reduced that strain. Longer Wu's also provide some measure of time to for the Server to recover to the Participants if something does go awry with the Server ... :) |
Send message Joined: 22 Mar 08 Posts: 7 Credit: 9,175,991 RAC: 0 ![]() ![]() |
You may find that a lot of people are unable to meet the deadline for these new WUs unless you increase it somewhat. It just gave me 20 WU's with a deadline of 5 days. Now I'm running a Q9450 @ 3.4GHz (WU currently @ 51mins and 10%. So even at 3 or 4 hrs per WU that's 60 - 80 hours of work to be done in 5 days, plus other work that I'm already committed to. I know I can finish them on time but, I don't think many others will be so lucky. Either increase the deadline or shorten the WUs a touch. Edit: It also means, as soon as the first WUs are finished, I won't be able to get new work from other projects for a few days, but I can live with that. |
![]() Send message Joined: 22 Feb 08 Posts: 260 Credit: 57,387,048 RAC: 0 ![]() ![]() |
I've increased the length of the WUs in an effort to ease the server traffic. I hope that this fixes a lot of the assimilator and traffic issues. Be sure to let me know if it works! As a heads up, I increased the time by increasing the accuracy with which we do the integral calculation over the wedge volume. It's possible that I made it too dramatic and increased the runtime by too much, so please let me know what completion times you are getting so I can adjust the next searches accordingly. just ran into some of those. Although not ready, it looks like 14-15hrs on a C2Q 6600. mic. |
![]() ![]() Send message Joined: 8 Apr 08 Posts: 45 Credit: 161,943,995 RAC: 0 ![]() ![]() |
6 units of GS_3720282 completed on Q9300 linux64 all between 148 and 150 min same units P4_3.0 & UBUNTU32 ... boinc view estimates at 10.5 hours/ unit OWN every thing I need EARN.. enough to live !!! WANT a solar array on the roof so I can run a BOINC farm( DREAM on!!) NO wife NO kids NO troubles |
Send message Joined: 2 Apr 08 Posts: 32 Credit: 1,017,362 RAC: 0 ![]() ![]() |
Got a bit of a problem that's stumping me. I awoke to find one MW run done at 6 hrs and 33 minutes...(Mac, Core 2 Duo @ 2.2 Ghz) so it updated and reported to server. However there is no record of this in my results or in pending credit. So I'm not sure if it was reported or lost in hyperspace. Secondly as someone else mentioned the deadlines need to be tweaked or the amount of work handed out per run should be tweaked...I've got running high priority on my one machine trying to meet the deadlines. But overall good idea to increase the WU's...hope you catch up on the technology end soon! |
![]() ![]() Send message Joined: 28 Feb 08 Posts: 2 Credit: 62,202 RAC: 0 ![]() ![]() |
Shouldnt you put a longer deadline than 5 days for the new units? Cause my computer downloaded 20 wus - believing they'ld last a few minutes... 20 wus @ several hours / 5 days = big problems ahead. ![]() |
![]() ![]() Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 ![]() ![]() |
Got a bit of a problem that's stumping me. I awoke to find one MW run done at 6 hrs and 33 minutes...(Mac, Core 2 Duo @ 2.2 Ghz) so it updated and reported to server. However there is no record of this in my results or in pending credit. So I'm not sure if it was reported or lost in hyperspace. WUs get quickly purged from the db, so that's probably why you can't see it any more. ![]() Join Support science! Joinc Team BOINC United now! |
![]() ![]() Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 ![]() ![]() |
Shouldnt you put a longer deadline than 5 days for the new units? I agree... 7 days should be a good setting.
The BOINC client will adjust its work request to match the size/run time of the new wus. You should be able a difference when you finished the first of the big ones. ![]() Join Support science! Joinc Team BOINC United now! |
Send message Joined: 3 Dec 07 Posts: 7 Credit: 1,001,407 RAC: 0 ![]() ![]() |
I think there is an issue with these WUs - my 1.8GH G5 Mac that crunched the old WUs in 12 minutes is now going to take 1500 minutes perWU. so 20 have been downloaded - 5 days to complete but it will take 20 days+ this Mac has 100% CPU allocated so its flat out 24/7 what to do? |
![]() ![]() Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 ![]() ![]() |
I think there is an issue with these WUs - my 1.8GH G5 Mac that crunched the old WUs in 12 minutes is now going to take 1500 minutes perWU. Just let it crunch ... the estimated 1500 min/WU (i guess that's what boinc shows you) are based on the assumption that the progress is increasing linear... but that's not the case. At about 50% wu completion the crunching speed increases by 3 to 5 times ... ![]() Join Support science! Joinc Team BOINC United now! |
Send message Joined: 2 Apr 08 Posts: 32 Credit: 1,017,362 RAC: 0 ![]() ![]() |
Got a bit of a problem that's stumping me. I awoke to find one MW run done at 6 hrs and 33 minutes...(Mac, Core 2 Duo @ 2.2 Ghz) so it updated and reported to server. However there is no record of this in my results or in pending credit. So I'm not sure if it was reported or lost in hyperspace. Within 2 minutes of reporting? That's a bit of a quick turn around... |
©2024 Astroinformatics Group