ECJ-INTEREST-L Archives

November 2005

ECJ-INTEREST-L@LISTSERV.GMU.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Sean Luke <[log in to unmask]>
Reply To:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Date:
Tue, 29 Nov 2005 15:11:05 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
On Nov 29, 2005, at 1:31 PM, I Jonyer wrote:

> I am using the distributed evaluation feature of ECJ. It appears that
> individuals are not distributed most optimally among the evaluation
> slave
> nodes. If, for instance, I specify (for the sake of argument) 109
> as the
> population size and I use 10 eval nodes, 9 will have 10 individuals to
> evaluate, and one will get 19 nodes. Is this true?

This should not be the case, but I may have done my math wrong.  We
can look into it.


> Also, I may have individuals that finish quickly with their assigned
> individuals (due to the characteristics of specific individuals),
> while
> others will take much longer.

We have a heavily revised version of the distribution code in-house
which handles exactly this (and also handles nodes dying and new
nodes coming on-line).  It should be available fairly soon.  If
you're hoping to test the code, you could contact myself and Liviu
Panait,  ([log in to unmask]).

Sean

ATOM RSS1 RSS2