[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

lisp for supercomputing

>Date: Mon, 28 Sep 92 20:49:04 +0100
>From: hmadorf@eso.org (Hans-Martin Adorf)
>To: info-mcl@cambridge.apple.com

>I know that neither MCL nor Allegor CL are running on the CM5, but
>the consistent slower performance (two different Fortran compilers,
>two different CL compilers) of Lisp w.r to Fortran worries me.

It's quite ridiculous to compare Lisp vs. Fortran as implemented
on Macintoshes and use that to pass judgement on a choice for 
establishing a policy for supercomputers!

If I were you I'd call Thinking Machines and ask them for benchmark
statistics comparing their *Lisp vs. their Fortran running on a CM.
I used to work at TMC (many years ago), and from all I've heard
recently their *Lisp is quite respectably fast compared with Fortran.

>Any clue how to improve the performance? If the situation prevails, I
>will find it difficult to recommend CL for supercomputing

Please check out the file optimization-techniques.sit.hqx available
by ftp from cambridge.apple.com in /pub/MCL2/contrib. It's a MS Word
document describing some speedup tricks.

Finally, please remember that there's more to a programming language
than speed alone. You wouldn't buy a car solely on the basis of its
top speed, would you? Most scientists need to explore, so you should factor
in Lisp's significant superiority for programming in an exploratory style.