Deadline to submit finished work?

Questions and Answers : Getting started : Deadline to submit finished work?
Message board moderation

To post messages, you must log in.

AuthorMessage
darkknight

Send message
Joined: 7 May 09
Posts: 5
Credit: 17,002
RAC: 0
United States
Message 892504 - Posted: 7 May 2009, 23:54:05 UTC

In various threads, I have seen people mention something about a due date for work. What's that all about?
ID: 892504 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 892507 - Posted: 8 May 2009, 0:02:06 UTC - in response to Message 892504.  

In various threads, I have seen people mention something about a due date for work. What's that all about?

Each WU is copied into exactly enough tasks to make certain the science gets done. However, some tasks are not returned, and there has to be a mechanism to ensure that the science gets done. So, each task has a deadline from the time it is assigned to a computer. If the task is not returned by the deadline, a replacement is generated and sent out. If the original is returned BEFORE the replacement, then the original will get credit if it validates. If it is returned AFTER the replacement, it is very unlikely to get credit (the only exception is when the validator is backlogged). These are soft deadlines in that the work may be accepted somewhat late, but the later the work is returned, the less likely that credit will be granted. For S@H tasks, (not AstroPulse) a 90MHZ Pentium could comfortably complete the tasks on time if it was left on almost all of the time.


BOINC WIKI
ID: 892507 · Report as offensive
darkknight

Send message
Joined: 7 May 09
Posts: 5
Credit: 17,002
RAC: 0
United States
Message 892879 - Posted: 9 May 2009, 0:27:05 UTC

My computer is a dual core system, 2.4ghz. I see that the stuff it is working on right now is called Astropulse. It estimates that it will take over 200 hours to complete. I am letting it process the data for 6 hours a day, will that work out all right?
ID: 892879 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 892903 - Posted: 9 May 2009, 1:24:30 UTC - in response to Message 892879.  

My computer is a dual core system, 2.4ghz. I see that the stuff it is working on right now is called Astropulse. It estimates that it will take over 200 hours to complete. I am letting it process the data for 6 hours a day, will that work out all right?

I am not completely certain. I do know the AP estimates are a bit off, but I cannot remember whether it is high or low.

If the estimate is right, it will take just over a month - which is, I believe, the deadline. However, it is unlikely that a computer is going to be able to finish the work in the 3 or 4 days that the estimate says the task will be late.


BOINC WIKI
ID: 892903 · Report as offensive
Aurora Borealis
Volunteer tester
Avatar

Send message
Joined: 14 Jan 01
Posts: 3075
Credit: 5,631,463
RAC: 0
Canada
Message 892927 - Posted: 9 May 2009, 2:36:39 UTC
Last modified: 9 May 2009, 2:43:47 UTC

The estimate is probably high. If you divide the hours done by percentage done it will give you a more accurate run time. For your system I should think that will be under 100 hrs or less than 17 days at 6 hrs/day. The deadlines for Astropulse are 30 days.

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470
ID: 892927 · Report as offensive
darkknight

Send message
Joined: 7 May 09
Posts: 5
Credit: 17,002
RAC: 0
United States
Message 892935 - Posted: 9 May 2009, 3:43:34 UTC

Okay, that sounds good. I was afraid you were going to say like, a week or something.
ID: 892935 · Report as offensive

Questions and Answers : Getting started : Deadline to submit finished work?


 
©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.