Message boards : ATLAS application : Why are Runtimes so Variable?
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile rbpeake

Send message
Joined: 17 Sep 04
Posts: 105
Credit: 32,824,862
RAC: 59
Message 50425 - Posted: 20 Jun 2024, 22:13:31 UTC

Assuming each work unit consists of 400 tasks, there are instances when the runtimes (on the same machine, no change in setup) vary widely.

For example, a typical runtime for my setup is approximately 26,000 seconds, but this can vary sometimes from 8,800 up to 32,000 seconds.

Thanks.
Regards,
Bob P.
ID: 50425 · Report as offensive     Reply Quote
greg_be

Send message
Joined: 28 Dec 08
Posts: 341
Credit: 4,865,275
RAC: 87
Message 50446 - Posted: 24 Jun 2024, 17:30:35 UTC - in response to Message 50425.  

All I can think of is the type of protons they are smashing together in virtuality.
Not all protons smash together equally.
Maybe you get more beams out of one than the other?
ID: 50446 · Report as offensive     Reply Quote

Message boards : ATLAS application : Why are Runtimes so Variable?


©2024 CERN