Advanced search

Message boards : Number crunching : NATHAN vs NOELIA

Author Message
Matt
Avatar
Send message
Joined: 11 Jan 13
Posts: 216
Credit: 846,538,252
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 31082 - Posted: 28 Jun 2013 | 3:22:38 UTC

I've noticed a difference in how the latest NOELIA and NATHAN WUs use resources and grant credit. I'm just curious what they are doing differently (i.e. research, CPU vs GPU intensive, etc.) and how this plays into the overall granting of credit. NATHAN seems to closely match GPU and CPU time. NOELIA had quite a disparity.

I'm not a credit junkie. I just noticed a difference and am curious what the difference is behind it.

For Reference:
NOELIA WU http://www.gpugrid.net/workunit.php?wuid=4549893
NATHAN WU http://www.gpugrid.net/result.php?resultid=6993967

Stefan
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 5 Mar 13
Posts: 348
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 31090 - Posted: 28 Jun 2013 | 12:46:41 UTC
Last modified: 28 Jun 2013 | 12:58:51 UTC

Nate is on holidays right now so I won't give you the best answer but I think it depends on two factors. The biological system in question (simulation box dimensions / number of atoms to simulate) and also if they use some CPU functions and how often they use them.
Before some months Noelia was working with some CPU bound functions so this caused her a slowdown, but at this point I think she doesn't use them anymore (I think we implemented them on GPU's) so it probably depends on the biological system.

As on credits, I don't personally know the formula with which they are calculated, but they cannot be done very exactly (apparently). So there will always be some (at least minor) disparity there.

Post to thread

Message boards : Number crunching : NATHAN vs NOELIA

//