conrad@uunet.UU.NET (Conrad Cunningham) (05/26/88)
Readers of this news group may be interested in a book recently
published. The book is:
Parallel Program Design: A Foundation
by K. Mani Chandy and Jayadev Misra
(both of the University of Texas--Austin)
published by Addison-Wesley in 1988
In the Forward C. A. R. Hoare says:
It is not often that we can welcome the advent of a new theory
of programming. Twelve years ago, E. W. Dijkstra published his
DISCIPLINE OF PROGRAMMING, which is still a definitive source-book on
the development of sequential algorithms. And now Chandy and Misra
have shown how Dijkstra's methods, and more recent developments, can
be generalized to distributed and concurrent algorithms. Their work
deserves the warmest welcome.
In the Preface the authors state:
The thesis of this book is that the unity of the programming
task transcends differences between the architectures on which
programs can be executed and the application domains from which
problems are drawn. Our goal is to show how programs can be developed
systematically for a variety of architectures and applications. The
foundation, upon which program development is based, is a simple
theory: a model of computation and an associated proof system.
In the book the authors present their philosophy of
programming, their UNITY programming notation and proof system, and
their methodologies for concurrent program development. They then
apply their theory to a variety of example "applications" and
"architectures". The discussion of examples probably makes up 2/3 of
the 500 pages of text.
Conrad Cunningham conrad@wucs1.wustl.edu
Department of Computer Science
Washington University in St. Louislim@cit-vax.Caltech.Edu (Kian-Tat Lim) (05/27/88)
I was fortunate enough to take a course with Dr. Chandy, who is visiting Caltech this year. We used a preprint of his book, and I was very impressed with its clarity and uniformity of presentation and its espousal of a machine-independent, stepwise-refinement approach to parallel programming. (Bias: this was the first formal course in concurrent programming I have taken.) Recommended reading. -- Kian-Tat Lim (ktl@wagvax.caltech.edu, GEnie: K.LIM1)
eugene@pioneer.arc.nasa.gov (Eugene N. Miya) (05/28/88)
>(Bias: this was the first formal course in concurrent programming I have >taken.) Recommended reading. Just a dissenting opinion. I was not overwhelmed by this book. I won't be putting a recommended keyword onto its entry unless there are a few more informed votes for it. UNITY, CSL, TRAC aren't really impressive in my book, but then, I've had difficulty getting TRs from Texas. Another gross generalization from --eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov resident cynic at the Rock of Ages Home for Retired Hackers: "Mailers?! HA!", "If my mail does not reach you, please accept my apology." {uunet,hplabs,ncar,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene "Send mail, avoid follow-ups. If enough, I'll summarize."