aglew@oberon.crhc.uiuc.edu (Andy Glew) (07/26/90)
This is another of those "Looking for ... programs for my research"
broadcast questions, with a twist: I have collected a number of
programs that I may be able to give others, or at least that I can
point to the authors of.
I am looking for parallel programs that I can use as subjects for
experiments in computer architecture (synchronization). Ideally these
would be written in C, and run on an Encore (or a Sequent). If they
use a non-vendor provided threads package, I'll need a pointer to
where I can get that threads package. (Explicit parallelism
preferred).
I already have (and am willing to provide to others, if it is okay
with the original authors):
gravsim
An N-body gravity simulator for galactic dynamics,
written in C, runs on Encore, Sequent, Alliant...
Written by mbellon@urbana.mcd.mot.com
locusroute
pthor
Unfortunately, I have not yet succeeded in
getting these to work on our Encore. My copies
are old and grotty. Can anyone point me to fresh
versions?
I have several others that I am not yet certain that the authors wish
to have distributed. I'll ask. If it's okay, I'll post.
I will respond to a reasonable number of email requests,
but if it becomes overwhelming I will get these put in UIUC's
anonymous ftp area, and then post where they are.
So, having exposed myself to lots of "Give me everything you've got"
requests, does anyone else out there have any other parallel programs
that they would be willing to let me take measurements, traces, etc.
of?
I am particularly interested in non-scientific applications, since
the biggest market for Sequent style applications seems to be
databases. Unfortunately, I cannot buy a commercial database, but is
there any public domain multiuser database code out there that uses
the client server model, or parallelizes queries, or....? Multiuser
games, military simulations - anything using a shared memory model.
Trusting in the net and serendipity...
--
Andy Glew, andy-glew@uiuc.edu
Propaganda:
UIUC runs the "ph" nameserver in conjunction with email. You can
reach me at many reasonable combinations of my name and nicknames,
including:
andrew-forsyth-glew@uiuc.edu
andy-glew@uiuc.edu
sticky-glue@uiuc.edu
and a few others. "ph" is a very nice thing which more USEnet
sites should use. UIUC has ph wired into email and whois (-h
garcon.cso.uiuc.edu). The nameserver and full documentation are
available for anonymous ftp from uxc.cso.uiuc.edu, in the net/qi
subdirectory.aglew@oberon.crhc.uiuc.edu (Andy Glew) (07/27/90)
In case it was not clear in the original post, I have locusroute and
pthor, but (1) am not sure if I can give them out, because I have not
yet managed to contact the original authors, and (2) it wouldn't be
much use to give them out, because the versions I have are not
working. I mentioned them only in the hopes that somebody else has
working versions that they are able to give to me.
I know that I can distribute gravsim (with the usual restrictions).
I have a few other programs that I am hoping to secure permission to
distribute.
--
Andy Glew, andy-glew@uiuc.edu
Propaganda:
UIUC runs the "ph" nameserver in conjunction with email. You can
reach me at many reasonable combinations of my name and nicknames,
including:
andrew-forsyth-glew@uiuc.edu
andy-glew@uiuc.edu
sticky-glue@uiuc.edu
and a few others. "ph" is a very nice thing which more USEnet
sites should use. UIUC has ph wired into email and whois (-h
garcon.cso.uiuc.edu). The nameserver and full documentation are
available for anonymous ftp from uxc.cso.uiuc.edu, in the net/qi
subdirectory.aglew@oberon.crhc.uiuc.edu (Andy Glew) (08/01/90)
I recently posted seeking parallel programs for use in research into
parallel computer architectures, offering one, gravsim, for use by
other researchers, and mentioning a number of others.
I have just emailed a tar-compress-uuencode-split-sharred gravsim to
the several people who requested it. If you do not receive this
either I'm having email problems, or I didn't get your request.
I have also learned of several other sources of parallel applications
that may be useful:
Stanford University is assembling a parallel benchmark suite, that
will include locusroute and pthor and others. It will soon be
publicized. Thus, waiting for this benchmark suite may be easier than
bothering the authors of these programs.
The DECUS public domain software library tapes apparently contain
a number of parallel programs written for VAX VMS with symmetric
multiprocessing. Notable among these is Cayenne, a parallel version
of SPICE. I have not yet obtained these, but they sound very useful.
Contact DECUS for membership and ordering info.
Finally, several people contacted me saying that they have lent their
parallel applications to other researchers in the past, but refuse to
do it any more because they have been "burnt" by the experience - the
other researchers have "trashed" the contributed codes and attacked
the authors of the contributed benchmarks.
This is a sad commentary on the lack of basic etiquette in the
research community.
--
Andy Glew, andy-glew@uiuc.edu
Propaganda:
UIUC runs the "ph" nameserver in conjunction with email. You can
reach me at many reasonable combinations of my name and nicknames,
including:
andrew-forsyth-glew@uiuc.edu
andy-glew@uiuc.edu
sticky-glue@uiuc.edu
and a few others. "ph" is a very nice thing which more USEnet
sites should use. UIUC has ph wired into email and whois (-h
garcon.cso.uiuc.edu). The nameserver and full documentation are
available for anonymous ftp from uxc.cso.uiuc.edu, in the net/qi
subdirectory.Thierry.Priol@irisa.fr (Thierry Priol) (08/01/90)
>From article <9932@hubcap.clemson.edu>, by aglew@oberon.crhc.uiuc.edu (Andy Glew): > I recently posted seeking parallel programs for use in research into > parallel computer architectures, offering one, gravsim, for use by > other researchers, and mentioning a number of others. > A parallel ray-tracing algorithm is available by anonymous FTP on : irisa.irisa.fr (131.254.2.3) in directory iPSC2/VM_pRAY -- Thierry PRIOL Phone: 99 36 20 00 IRISA / INRIA U.R. Rennes Fax: 99 38 38 32 Campus Universitaire de Beaulieu Telex: UNIRISA 950 473F 35042 RENNES CEDEX - FRANCE E-mail: priol@irisa.fr
pcg@cs.aber.ac.uk (Piercarlo Grandi) (08/04/90)
"aglew" == Andy Glew writes:
aglew> Finally, several people contacted me saying that they have lent their
aglew> parallel applications to other researchers in the past, but refuse to
aglew> do it any more because they have been "burnt" by the experience - the
aglew> other researchers have "trashed" the contributed codes and attacked
aglew> the authors of the contributed benchmarks.
aglew> This is a sad commentary on the lack of basic etiquette in the
aglew> research community.
You mean the fact that researchers can be just as reluctant as
commercial concerns to subject their work to public scrutiny, because
they fear that the often poor quality and abundance of mistakes therein
would put their results in question and harm their image?
This would be understandable, after all researchers have to pay a
mortgage too. This is far more important than peer review and the
admittedly fairly recent (just about 300 years) idea that independent
criticism and repeatability of experiment are the cornerstones of
scientific advancement. Only people that have already made it (like the
late Feynman) are antisocial enough not to worry about the mortgage or
the shareholders.
Too bad that the constant debate about quality of benchmarking in this
newsgroup and pitfalls of benchmarks is stupidly based on such old
delusions. A lot of trouble would be saved if we stopped being nosey
about benchmark claims, and the real sadness is that this has not
happened yet.
--
Piercarlo "Peter" Grandi | ARPA: pcg%cs.aber.ac.uk@nsfnet-relay.ac.uk
Dept of CS, UCW Aberystwyth | UUCP: ...!mcsun!ukc!aber-cs!pcg
Penglais, Aberystwyth SY23 3BZ, UK | INET: pcg@cs.aber.ac.uk