lengauer@lfcs.edinburgh.ac.uk (Chris Lengauer) (02/15/91)
This is in response to your inquiry on comp.parallel.
I have been working on the subject of loop parallelization
in the systolic paradigm. The target is occam-like code,
not data flow code. Here are some papers (in BIBTEX format):
author="C.~Lengauer",
title="Code Generation for a Systolic Computer",
journal="SOFTWARE---Practice and Experience",
volume="20",
number="3",
month="Mar.",
year="1990",
pages="\mbox{261--282}",
author="C.~Lengauer and D.~G.~Hudson",
title="A Systolic Program for {Gauss-Jordan} Elimination",
booktitle="Beauty Is Our Business",
editor="W.~H.~J. Feijen and A.~J.~M. van Gasteren, D.~Gries and J.~Misra",
series="Texts and Monographs in Computer Science",
publisher="Springer-Verlag",
year="1990",
chapter="30",
pages="\mbox{262--273}",
author="C.~Lengauer and M.~Barnett and D.~G.~Hudson",
title="Towards Systolizing Compilation",
journal="Distributed Computing",
volume="5",
number="1",
year="1991",
note="To appear",
author="M.~Barnett and C.~Lengauer",
title="A Systolizing Compilation Scheme",
institution="Department of Computer Sciences,
The University of Texas at Austin",
month="Jan.",
number="TR-91-03",
year="1991",
The second paper is a published subset of the third; the third paper
is still in press.
To obtain the last paper, please send a request to Mike Barnett at:
mbarnett@cs.utexas.edu.
Another person you might want to get in touch with is Duncan Hudson
at UTexas. His e-mail address: dghudson@cs.utexas.edu.
I'd be interested in your compilation of responses.
Chris Lengauer.