Re: About the performance of named references / non-logical variable

From: Joachim Schimpf <j.schimpf_at_icparc.ic.ac.uk>
Date: Wed 02 Mar 2005 02:16:06 PM GMT
Message-ID: <4225CAA6.8080104@icparc.ic.ac.uk>
Ulrich Scholz wrote:
> Hi everybody,
> 
> I wonder about the performance of using a named reference compared of using
> an additional argument of a predicate.

Passing an extra argument adds a small overhead in many places.
Accessing a named reference adds a larger overhead in fewer places.
So, as usual, it depends on your application...


 > And how about a non-logical variable?  Are they much slower?

Named references and non-logical variables are not interchangeable
except in special circumstances (see User Manual section on global
references).  For non-logical variables the speed depends on the size
of the term that is being set/gotten, which means that they are
slower when that term is not atomic.


-- 
  Joachim Schimpf              /             phone: +44 20 7594 8187
  IC-Parc                     /      mailto:J.Schimpf@imperial.ac.uk
  Imperial College London    /    http://www.icparc.ic.ac.uk/eclipse
Received on Wed Mar 02 14:22:24 2005

This archive was generated by hypermail 2.1.8 : Wed 16 Nov 2005 06:07:34 PM GMT GMT