teraFLOPS estimate?

Message boards : Number crunching : teraFLOPS estimate?

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile [HWU]Flotta Stellare - Starfleet

Send message
Joined: 20 Jul 06
Posts: 8
Credit: 10,549,926
RAC: 0
Message 55055 - Posted: 12 Aug 2008, 17:23:48 UTC

79teraflops now, good good
ID: 55055 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Michael G.R.

Send message
Joined: 11 Nov 05
Posts: 264
Credit: 11,246,222
RAC: 14
Message 55059 - Posted: 12 Aug 2008, 21:05:36 UTC
Last modified: 12 Aug 2008, 21:06:05 UTC

I wonder what causes these long plateaus followed by quick jumps.

It is the metric that isn't accurate, or is it the actual processing power that follows these cycles?

Is it a new wave of quad-cores that have been bought before the next school year that are coming online? Some other big project is out of work?

Anyone has a theory?
ID: 55059 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Mod.Sense
Volunteer moderator

Send message
Joined: 22 Aug 06
Posts: 4018
Credit: 0
RAC: 0
Message 55064 - Posted: 13 Aug 2008, 1:43:15 UTC

The file server was down for a period of time, which prevented credit from being issued for a while. Since most machines still had work and were crunching it, there was a surge of reorts of completed work when the system came back online. This causes day to day swings.

A plateau, followed by a jump which is sustained over time is often due to teams moving resources to their project of the month, or due to press releases mentioning the project.
Rosetta Moderator: Mod.Sense
ID: 55064 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 · 2

Message boards : Number crunching : teraFLOPS estimate?



©2021 University of Washington
https://www.bakerlab.org