Message boards :
Sixtrack Application :
No more workspace1_HEL_Qp_2_MO_0_1t_3s_1e7turns_BOINC tasks for my small laptop
Message board moderation
Author | Message |
---|---|
Send message Joined: 4 Apr 19 Posts: 31 Credit: 4,423,972 RAC: 0 ![]() ![]() |
Please don't send me these tasks with deadline of just 7 days. I just completed my first 1e7turns task with more than 96 hours of processing by turnning on my computer when it should be powered off, and allowing it to use single core boosting which is not productive overall. I'm completing my second 1e7turns task now. No new tasks. |
Send message Joined: 29 Feb 16 Posts: 157 Credit: 2,659,975 RAC: 0 ![]() ![]() |
Hi Sesson, apologies for that. Yes, you are right, the deadline of 7 days is too tight for this kind of studies, and at submission this parameter was left behind. Please feel free to abort any of these tasks or temporarily detach your laptop from SixTrack for let's say a week. Thanks for your understanding, Cheers, A. |
![]() ![]() Send message Joined: 29 Sep 04 Posts: 281 Credit: 11,866,264 RAC: 0 ![]() ![]() |
If the 10^7 turn jobs were still a "test application" then Sesson, or others, could uncheck that in their preferences, while still being able to run "ordinary" Sixtracks. I don't know if that is a distinction that is possible on the project side? |
Send message Joined: 29 Feb 16 Posts: 157 Credit: 2,659,975 RAC: 0 ![]() ![]() |
Hi Ray, yes you are right. But this batch of 10^7 turns jobs were submitted (wrongly) to sixtrack instead of sixtracktest, with the default of 1w of delay bound. On sixtracktest we have presently 2w. But it would have been not respectful towards those users with hosts capable of crunching the tasks on time which have already started crunching them, or even producing results. Hence, I let them go... I will remind the users to load sixtracktest and not sixtrack for this kind of studies. Happy crunching, Cheers, A. |
![]() ![]() Send message Joined: 29 Sep 04 Posts: 281 Credit: 11,866,264 RAC: 0 ![]() ![]() |
Thanks, Alessio, I did suspect that there would be no way to alter that once they had been released into the wild. Longest completed one so far for me was 42.7hrs with another of 33hrs. Most others have not been unusually long. 1 running at 83% 80hrs (sse2 Linux64) finished after 97hrs 1 running at 68% 69hrs (win32 on 64bit host) finished after 81hrs 1 running at 56% 75hrs (win32 on 64bit host), finished after 132hrs (5+1/2days) I have no reason to doubt that these will also finish successfully. I'm happy to do as many of these as arrive as my machines mostly run 24/7 (1 gets reduced to 2 cores overnight as it has a noisy fan when flat out) but I keep my buffers at 0.2 days so as not to get accidentally swamped. Update edits in bold |
![]() ![]() Send message Joined: 24 Oct 04 Posts: 1184 Credit: 57,287,957 RAC: 63,172 ![]() ![]() |
I have run lots of these long Sixtrack workspace1_HEL_Qp_2_MO tasks and some as long as 90 hours and don't mind doing them as long as all the Wings out there don't just abort them when they see how long they can be. The main thing to do is make sure your Prefs are set so you don't get more tasks than you can run at the same time as far as your cores since setting at No Limit and getting more than you can possibly run before due time you will either have to Abort or wait for the server to take them back which doesn't help members that did already run and turn them in. |
©2025 CERN