Considerations for Massively Parallel Unix Systems on the Nyu Ultracomputer and Ibm Rp3 (Classic Reprint) - Tapa dura
Sinopsis
Excerpt from Considerations for Massively Parallel Unix Systems on the Nyu Ultracomputer and Ibm Rp3
However, it remains to be demonstrated that such machines can be effectively utilized, and that unix is well-suited to the needs of such an environment. There are two aspects to this challenge. First, several thousand processors must be coordinated in such a way that their aggregate power is applied to useful computation. Serial code sections in which one processor works while the others wait become bottlenecks that drastically reduce the power obtained, even if the serial section is so small or infrequently executed as to be entirely acceptable on a machine with only modest parallelism. Indeed, the relative cost of a serial bottleneck rises linearly with the number of processors involved. Second, the machine must be programmable by humans. Effective use of high degrees of parallelism will be facilitated by simple languages and facilities for designing, writing, and debugging parallel programs.
About the Publisher
Forgotten Books publishes hundreds of thousands of rare and classic books. Find more at www.forgottenbooks.com
This book is a reproduction of an important historical work. Forgotten Books uses state-of-the-art technology to digitally reconstruct the work, preserving the original format whilst repairing imperfections present in the aged copy. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in our edition. We do, however, repair the vast majority of imperfections successfully; any imperfections that remain are intentionally left to preserve the state of such historical works.
"Sinopsis" puede pertenecer a otra edición de este libro.
Reseña del editor
Excerpt from Considerations for Massively Parallel Unix Systems on the Nyu Ultracomputer and Ibm Rp3
When concurrent loads, stores, and fetch-and-adds are directed at the same memory location and meet at a switch, they can be combined without introducing any delay (see Klappholz and Gottlieb et at. Since combined requests can themselves be combined, any number of concurrent memory references to the same location can be satisfied in the time required for one central memory access. It is this property that permits the bottleneck-free implementation of many coordination protocols.
The impact of network latency on performance is reduced by associating a local cache memory with each pe. Frequently-used program code and data can be accessed in (approximately) a single processor cycle when resident in the cache. Thus the network latency is eliminated from many memory accesses, and all accesses benefit from the reduced network traffic.
About the Publisher
Forgotten Books publishes hundreds of thousands of rare and classic books. Find more at www.forgottenbooks.com
This book is a reproduction of an important historical work. Forgotten Books uses state-of-the-art technology to digitally reconstruct the work, preserving the original format whilst repairing imperfections present in the aged copy. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in our edition. We do, however, repair the vast majority of imperfections successfully; any imperfections that remain are intentionally left to preserve the state of such historical works.
"Sobre este título" puede pertenecer a otra edición de este libro.
