0 Credits

Emil
Emil
Joined: 10 Dec 11
Posts: 2
Credit: 2327
RAC: 0
Topic 196111

I am a newbee here, so be patient :)
I run my processor on 2 projects now, SETI@Home and Einstein@Home. When work's done, my credits in SETI@Home are up, while it's still 0 in Einstein@Home, even if I have now made a lot of tasks. Can you mention what may be causing that?

Jord
Joined: 26 Jan 05
Posts: 2952
Credit: 5893653
RAC: 1

0 Credits

In the mean time, your credits are off that zero.
All work at Seti and Einstein is sent to at least two computers, which both have to return same results (checked by the validator), before credit is given out. When work is not returned within the deadline limit, or when either result isn't matching with the other, the task will be sent to a third computer, or a next, until a matching result has been returned and credit can be granted.

Just returning work doesn't mean you'll get credit right away.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5877
Credit: 118650087209
RAC: 18832901

RE: Just returning work

Quote:
Just returning work doesn't mean you'll get credit right away.


Well, actually it does - in Emil's case :-).

This is the list of tasks for the host in question. Two tasks were downloaded on Dec 10 and two on Dec 11. Emil says he has "made a lot of tasks", but only two have been returned as of now. They were both returned on Dec 18 and were immediately credited since the two wingmen involved had returned their tasks on Dec 11, a whole week earlier. So, instant credit.

The next two tasks are still 'in progress'. They too (when returned) should be instant credit because the next two wingmen have also returned their tasks many days ago. So it's not the case that Emil is being held up for credit, but rather the other way around :-). Maybe we will get complaints from Emil's wingmen :-).

One final comment for Emil. Your dual core host can process FGRP tasks in around 11 hours, which is quite reasonable. Your first tasks were returned after 8 days, which seems like a very long time. Perhaps you have a very small resource share for Einstein and a very large one for Seti. This is quite OK if you wish but it would explain why you have to wait so long to see your E@H credit. BOINC might allocate 1 hour to E@H but then have to spend many hours on Seti in order to honour your resource shares. If this is so, you may have problems if Seti is ever unable to supply work. You may find BOINC will fill your cache with E@H tasks and then later on (when Seti has more work) will have to run them in panic mode so as not to miss the deadline. One way to protect against this is not to have an excessive work cache.

Cheers,
Gary.

Jord
Joined: 26 Jan 05
Posts: 2952
Credit: 5893653
RAC: 1

RE: Well, actually it does

Quote:
Well, actually it does - in Emil's case :-)


Please don't confuse the user, who may think he's special and now always gets credit like that. Getting credit isn't just returning work, it's dependent on what your wingman did with the task as well:
- did he return it already;
- is he ever going to return it;
- is his cache one of a couple of tens or a couple of thousands;
- is he returning correct work or does it need to be resent to a third computer;

And the most elusive of all:
- will the work validate at all?

Emil
Emil
Joined: 10 Dec 11
Posts: 2
Credit: 2327
RAC: 0

Thanks! At least I won't be

Thanks! At least I won't be afraid now that my CPU Time for Einstein@Home will be lost. And I learned something about how project works.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.