Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site ccice2.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!godot!harvard!seismo!rochester!ritcv!ccice5!ccice2!bwm
From: bwm@ccice2.UUCP (Brad Miller)
Newsgroups: net.lang.lisp,net.lang.prolog
Subject: Wanted: info about lisp/prolog performance
Message-ID: <518@ccice2.UUCP>
Date: Wed, 5-Dec-84 19:00:17 EST
Article-I.D.: ccice2.518
Posted: Wed Dec  5 19:00:17 1984
Date-Received: Sat, 8-Dec-84 05:24:55 EST
Distribution: net
Organization: CCI Central Engineering, Rochester, NY
Lines: 19
Xref: watmath net.lang.lisp:283 net.lang.prolog:427

I am looking at performance in lisp and prolog: specifically, what parts of
the language (or their typical implementations) consume the most resource in
typical programs, e.g. language parsers, etc.

Does anyone have any info on this, or good pointers? I have found some info
'indirectly', i.e. by what others have improved on their systems to improve
performance, but I wonder if there is some definitive work (or set of works)
on the subject? Perhaps someone has already done some performance work on their
own?

I am fairly well aware of the 'typical' things: i.e. garbage collection efficiency,
searching the OBLIST, etc. I suppose I am more interested in algorithms that have
proven useful (like H. Lieberman and C. Hewitt's paper on "A Real-Time Garbage
Collector Based on the Lifetimes of Objects" in CACM) and hardware speedups.

Thanks for any info!
Brad Miller
-- 
...[rochester, cbrma, rlgvax, ritcv]!ccice5!ccice2!bwm