[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: execution time
From: |
Marcus G. Daniels |
Subject: |
Re: execution time |
Date: |
05 Dec 2000 23:29:41 -0800 |
User-agent: |
Gnus/5.070084 (Pterodactyl Gnus v0.84) Emacs/20.4 |
>>>>> "SR" == M Lang / S Railsback <address@hidden> writes:
SR> Let me ask a related question- what might be a good hardware
SR> solution to the problem of models with lots of agents doing lots
SR> of things, assuming no parallelization?
I can't say I've done any low-level cache profiling or anything like that.
PCs tend to have large caches, so they should do pretty good as far as
memory speed goes.
SR> Under what conditions is it a big advantage to have two
SR> processors? I suppose multiple model runs can be shipped to
SR> different processors- does a workstation running NT do stuff like
SR> that automatically?
A symmetric multiprocessing system (SMP), or a multi-machine cluster runing
Mosix, will let you spawn N model runs with different parameters and
keep all the processrs busy. But having N processors won't help you
with any single simulation run. NT does do SMP, but doesn't do it cross-box
like Mosix, you've got to have multiple processors in the same box.
==================================
Swarm-Support is for discussion of the technical details of the day
to day usage of Swarm. For list administration needs (esp.
[un]subscribing), please send a message to <address@hidden>
with "help" in the body of the message.