The Connection Machine (Artificial Intelligence)

The Connection Machine describes a fundamentally different kind of computer. It offers a preview of a parallel processing computer that Daniel Hillis and others are now developing to perform tasks that no conventional, sequential machine can solve in a reasonable time.

W. Daniel Hillis is a founder of Thinking Machines Corporation where he is engaged in building connection machines as a significant step toward real thinking machines. The Connection Machine is included in the Artificial Intelligence series, edited by Patrick Winston, Michael Brady, and Daniel Bobrow.

This book is essentially an edited version of Hillis's landmark thesis describing the design and implementation of the Connection Machine (CM), a massively parallel computer. The philosophy behind the CM's design is that the right kind of machine for many important computational tasks is a machine with vast numbers of simple processors doing the same thing on different data. This notion of one processor per important data element (one processor per pixel in image processing) is inspiring.

The Connection Machine is not a textbook and may be intimidating to beginners, but it provides a wonderful picture of the kinds of issues involved in designing a new machine. The book is well written and features a host of interesting discussions by Hillis on related topics (such as general philosophy of parallel computing). Anyone interested in the subject of computer architecture will enjoy and profit greatly from this book.