Author: Robert Hyatt
Date: 19:31:26 09/15/01
Go up one level in this thread
On September 15, 2001 at 05:57:28, Dan Andersson wrote: >Due to the latencies incurred when communicating in a network. You need >different work distrubution algorithms when programming a cluster compared to a >SMP machine. There exists more than a few such algorithms that will get near >linear speedups until the communication ceiling is reached. Not in chess I am afraid. IE I am not aware of any algorithm that will produce linear speedups on a shared memory machine, assuming by "linear" you mean a line with slope of 45 degrees (ie 16 cpus is 16 times faster than 1 cpu.) a slope of 3/4 is considered very good.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.