Customer Reviews

Parallel Programming in C with MPI and OpenMP

5 star:		(4)
4 star:		(1)
3 star:		(2)
2 star:		(1)
1 star:		(0)

 

 

I used this book in my parallel computing course. It is really a good introductory book on parallel programming. Some examples are given in the book. The first seven chapters discuss parallel algorithm design, MPI, examples on The sieve of Eratosthenes and Floyd's algorithm, and performance analysis. Some other topics that also covered (some of these may or may not have working examples) include the matrix-vector multiplication, Monte Carlo, matrix multiplication, solving linear systems, sorting, FFT, search, and finite difference methods. Since there are so many topics to be covered in a book of only 500 pages, the discussions on most of these topics are not very depth. Overall, I still recommend this book for beginners in parallel algorithm design.

The thing that impressed me was that the writing was so clear. True, the sentences tend to be short, but that is high virtue in technical writing. My students have been very positive about the book.

I also think the mathematical analysis is good, too, not too easy but not super hard either.

This book just pulls together all the crucial information between two covers. I find myself agreeing almost exactly with where he places his "key" symbols in the margins to highlight important sentences, which is also a good sign that the book is "right on."

Well, to begin with, for a book that has "...programming in C..." in the title, there is very little C code in the text-- most is pseudocode. Most of the book is an analysis of various parallel algorithms, with very little instruction on how to use MPI. There are much better resources out there for learning MPI, as Quinn only covers about 30 of the over 100 functions in MPI, without all that much detail. Some of extremely important and necessary concepts of parallel programming are only mentioned in passing...such as load balancing. Most of the applications (with the exception of matrix operations) are simple and basic to the point of making me wonder why you would even bother parallelizing them...even as a pedagogical tool. The book does a very good job of analyzing algorithms, but calling it an "introduction to MPI" or even an "introduction to parallel programming" textbook is incorrect.

This book is a great introduction to the theory of parallel programming. It is important to note that it is not a great reference for MPI, but it does a good job introducing the basic MPI functions and how to implement parallel programs using them.

If you are looking for a good parallel programming primer, this book is a good start. If you are looking for an MPI reference guide, or a detailed discussion of more advanced MPI use, look elsewhere.

I used this as a textbook for a parallel programming course in 2005. The author goes into a fair amount of detail about a number of different algorithms (e.g., fast fourier transforms, differential equations, sorting, alpha-beta pruning) rather than focusing exclusively on parallel programming constructs. I consider this a feature; the algorithms serve as good motivation and illustrations of the parallel programming concepts that are presented.

The summaries of the MPI commands in the appendix are as good as anything I've found on the web.

The book also gives detailed examples (code) of how to do mundane things like distributing the contents of a file across distributed memory processors and using your random number generator in such a way as to guarantee that your program produces the same results irrespective of the number of processors it runs on.

This book provides very good introductory material to beginners of parallel programming in MPI. It provides many examples and through them one can learn not only the syntax of MPI but also how to design parallel programs.

If you're looking for a textbook for a class on parallel programming, this may be for you, but I wouldn't know. If you're trying to figure out MPI to actually write some code, this one definitely isn't for you. Get Pacheco's book instead. Half the price, too.

Basically, there is very little tutorial-style programming coverage in here, more design and analysis of parallel algorithms. The majority of MPI function calls is not even covered in detail, but only listed in the appendix. Give me a break.

The shipping was very fast and the book is in good condition, except
for a lot of handwritten markings on all pages :(