Message boards :
Number crunching :
Why are we waiting?
Message board moderation
Author | Message |
---|---|
Send message Joined: 12 Oct 19 Posts: 3 Credit: 251,147 RAC: 0 |
Some time ago I noticed that it was taking an eternity for some of my WUs that I had run and returned to be validated, so I decided to stop for a while in favour of more immediate work. So it's a month later and I have just looked at my tasks and there is no change! I have a paltry 4 tasks outstanding but this is a principle thing, and I have 2 questions really which can be encapsulated by Workunit 1362205 Why do you wait so long to send them to people? This was sent to me 17th May and I returned it a week later. It was also sent out at the same time to someone else who aborted it on the same day which is fair enough, but it then sat for two months until the 17th July before being issued again!!! Why? To me it seems that work that can be validated should be prioritised over work that nobody has started. Why to you give such a long period of time to people to carry out these tasks? This same task above, reissued on the 17th July has a deadline of October 16th! Really? 3 months? Is that really necessary? I am going to take a wild stab in the dark here (to make a point) and guess that it is not outside the bounds of possibility that come October 16th that job will be found to have not been completed, and after a further 2 months, it will get reissued with a 3 month deadline taking me to March of the following year - literally 10 months after I completed my half of the work. Seriously do you guys there think this is reasonable? Or am I missing something fundamental about the nature of the project that means the work needs to be carried out a few months apart to minimise any possibility the project being affected by quantum entanglement! ;-) Just looking at another job - Workunit 1413899 - Issued 22 July, completed by me on the 24th, but sadly my fellow worker got an error while computing on the 24th July, but it still hasn't been sent out again. Seriously? Why not? I believe that If you prioritise the issue of WUs where the work that has been partly completed over completely unprocessed WUs, and shorten that deadline time significantly, then I feel sure more people will want to take part in this project and you will get a much better return rate of completed results, which has to be a good thing for both project and crunchers. |
Send message Joined: 7 Nov 19 Posts: 31 Credit: 4,245,903 RAC: 0 |
|
Send message Joined: 12 Oct 19 Posts: 3 Credit: 251,147 RAC: 0 |
https://quchempedia.univ-angers.fr/athome/forum_thread.php?id=109 Thanks for that Luigi - it's good to know people are looking at these - I am not quite sure why the admin hasn't responded to my thread though. In yours he says "... and on my end, it's about the scientific results I'm waiting for... At the moment, project is running and we don't plan to shutdown it." It seems to me that if you send out the work quicker to people, who are not getting disillusioned with the rate they are sent out, then he would get his results a whole lot quicker! And while English may well not be his native language, he should try a little harder not to accuse his interested and concerned-enough-to-write volunteers of "generalising" and "whining"! If I have to wait months for work I have done to be issued to someone for validation (never mind that it might be months before it gets done, or god forbid, reissued), then I am going to whine about it! I feel he he is not listening to us and sadly until he does and responds in a positive way, I for one shan't be doing any more on this project, good science or not. |
Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0 |
No, English is not my native language and my words probably go beyond my thoughts. Sorry about that. The current situation is complicated. We don't have an engineer on the project and I'm the only one doing everything on the Boinc part. I am a lecturer and researcher, plus the SARS-CoV-2 pandemic has added a lot of work and constraints to my activities. The wingman part is managed by a software brick I don't have control over. Changing the server settings in production is very risky, but I'll have to deal with it. Of course I remain silent on the forum, but I use my time to the maximum to improve the situation. I hope that my changes will have visible and positive results. |
Send message Joined: 12 Oct 19 Posts: 3 Credit: 251,147 RAC: 0 |
Excellent, and apology totally accepted. For the sake of the sanity of the volunteers running the tasks on this project, and in turn the project itself, I genuinely feel that this software brick you refer to needs some changes. I look forward to any progress. Thank you very much for getting back to me. |
Send message Joined: 30 Aug 20 Posts: 1 Credit: 0 RAC: 0 |
damotbe, Universe@home had a similar problem with getting resends out in a timely manor. It took a while but the project did eventually get if fixed. You might want to ask the project administrator over there what he did to fix resolve the issue. Good Luck DadX |
Send message Joined: 23 Jul 19 Posts: 289 Credit: 464,119,561 RAC: 0 |
Thank you for this information. I have a similar mechanism in place, but from what I understand it takes time for the priority system to become operational. I have also read how they fix the problem and my settings are inspired by it. Wait and see... |
Send message Joined: 13 Sep 19 Posts: 69 Credit: 399,347 RAC: 0 |
The current situation is complicated. We don't have an engineer on the project and I'm the only one doing everything on the Boinc part. It's a pity. A very little team of 2 or 3 people can do a lot of difference in a boinc project! |
©2024 Benoit DA MOTA - LERIA, University of Angers, France