Excerpt from Wireability of an Ultracomputer
An n stage omega network (lawrie 3, Coke and Lipovsky 1) can be laid out utilizing many different topologically equivalent interconnection networks. One such is a network utilizing a perfect shuffle interconnection between each stage. While this yields a network which uses a single fixed wiring pattern between each of its stages, the wiring required becomes excessive as the number of stages increases. Instead we propose to use a variant of the baseline network (wu and Feng 5) shown in Figure 1, which involves a series of different interconnection patterns in successive stages but which leads to a more modular and manageable design.
A major advantage of the network shown below is that it can be split in the middle, yielding two symmetric halves. These halves can then be split in half again and then rejoined, using a two dimensional interconnection scheme. This results in a total layout involving only one challenging backplane, but very short, straight wires at all other points of connection.
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.
Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.
Anbieter: Forgotten Books, London, Vereinigtes Königreich
Paperback. Zustand: New. Print on Demand. This book describes how technology will allow for an increase in the size of an ultracomputer to 4096 processors, and provides a plan for structuring the design and building it. Ultracomputers are shared -memory MIMD parallel machines with thousands of processing elements, and are wired using an Omega connection network. The author builds upon previous research and designs to show how the hardware can be structured using smaller subsets of processors and switches. The design for the 4096 processor ultracomputer is based on present -day wiring technology and electrical power considerations. The significance of this work is that it does not assume any new breakthroughs in LSI or interconnection technology. Everything presented in the book can be implemented with todayâs technology, construed conservatively. This book is a reproduction of an important historical work, digitally reconstructed using state-of-the-art technology to preserve the original format. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in the book. print-on-demand item. Bestandsnummer des Verkäufers 9781332212682_0
Anzahl: Mehr als 20 verfügbar
Anbieter: PBShop.store US, Wood Dale, IL, USA
PAP. Zustand: New. New Book. Shipped from UK. Established seller since 2000. Bestandsnummer des Verkäufers LW-9781332212682
Anbieter: PBShop.store UK, Fairford, GLOS, Vereinigtes Königreich
PAP. Zustand: New. New Book. Shipped from UK. Established seller since 2000. Bestandsnummer des Verkäufers LW-9781332212682
Anzahl: 15 verfügbar