Volume 7, Issue 1 -
Spring/Summer 1999

Volume 6, Issue 3
Fall 1998

Volume 6, Issue 2
Spring/Summer 1998

Volume 6, Issue 1
Winter 1998

Volume 5, Issue 4
Fall 1997

Volume 5, Issue 3
Summer 1997

Volume 5, Issue 2
Spring 1997

Volume 5, Issue 1
Winter 1997

Volume 4, Issue 4
Fall 1996

Volume 4, Issue 3
Summer 1996

Volume 4, Issue 2
Spring 1996

Volume 4, Issue 1
Winter 1996

Volume 3, Issue 4
Fall 1995

Volume 3, Issue 3
Summer 1995

Volume 3, Issue 2
Spring 1995

Volume 3, Issue 1
January 1995

Volume 2, Issue 4
October 1994

Volume 2, Issue 3
July 1994

Volume 2, Issue 2
April 1994

Volume 2, Issue 1
January 1994

Volume 1, Issue 4
October 1993

Volume 1, Issue 3
July 1993

Volume 1, Issue 2
April 1993

Volume 1, Issue 1
January 1993


CRPC's Paul Messina, in conjunction with Thomas Sterling and Paul H. Smith, has recently written a book on the development of high- performance computing as projected into the next two decades, released in August 1995.

Published by MIT Press, Enabling Technologies for Petaflops Computing discusses what goes into building a computer ten times more powerful than all the networked computing capability in the United States. It summarizes the history of petaflops research and describes the technical and policy agendas for what could be a 20-year effort to build a petaflops-scale machine.

Petaflops refers to the ability of a computer to calculate 10 ** 15 floating point operations per second. To fathom the speed of such a computer, one must consider that the most powerful computers in existence today are four thousand times slower and the fastest workstations are about 5 million times slower than a petaflops computer could be theoretically.

Chapters focus on four interrelated areas: applications and algorithms, device technology, architecture and systems, and software technology. The authors project that construction of an effective petaflops computing system will indeed be feasible in two decades, but effectiveness and applicability will depend on dramatic cost reductions in component costs, as well as innovative approaches to system software and programming methodologies. They stipulate that a mix of technologies, such as semiconductors, optics, and possibly cryogenics will be required. They also stress that while no fundamental paradigm shift in system architecture is expected, active latency management will be essential, requiring a high degree of fine-grain parallelism and the mechanisms to exploit it.

"The potential new device technologies that might be used in a petaflops computer are very exciting," says Messina. "I find it particularly interesting that in all probability, a mix of new and current technologies will be used for such a system. Even more fascinating are the new applications we've identified that might be enabled by systems with a petabyte of memory and a petaflops of computing speed."

Released in paperback, the 200-page Enabling Technologies for Petaflops Computing is listed at $26.95. It is available at bookstores or directly from MIT Press. To order, call (800) 356-0343 or (617) 625-8569. Electronic orders may be sent to mitpress-orders@mit.edu .

CRPC's Paul Messina, co-author of a newly released book on petaflops research and technology, is director of Caltech Concurrent Supercomputing facilities and assistant vice president for Scientific Computing, Caltech; and manager of High Performance Computing, Jet Propulsion Laboratory.

Table of Contents