Hosted with the Supercomputer Software Department RAS Home
Parallel Computing Technologies Third International Conference (PaCT-95) September 12-15, 1995. St.-Petersburg, Russia Proceedings Home PaCTs Prev Contents Next Authors
	Vol. 964

Standard Microprocessors Versus Custom Processing Elements for Massively Parallel Architectures

by D. Etiemble, C. Germain

Abstract:
Choosing a standard microprocessor or a custom processing element as CPU of a massively parallel architecture has major impact on the hardware and software costs. The standard microprocessors have a high performance/cost ratio, but the associated cache hierarchy leads to completely reconsider the programming of the applications that have been developed for vector supercomputers. Low hardware cost is associated to a high software development cost. New approaches, as multithreaded architectures, have low software costs, but expensive hardware costs.

Keywords: Hardware and Architecture



Classification:
	C.1	processor architectures
	C.2	computer-communication networks
	C.3	special-purpose and application-based systems; microprocessor/microcomputer applications; process control systems; real-time systems; signal processing systems
	C.4	performance of systems; design studies; measurement techniques; modeling techniques; performance attributes; reliability, availability, and serviceability
	D.1	programming techniques
	D.2	software engineering
	D.3	programming languages
	D.4	operating systems
	F.1	computation by abstract devices
	F.2	analysis of algorithms and problem complexity


Source:
D. Etiemble, C. Germain, Standard Microprocessors Versus Custom Processing Elements for Massively Parallel Architectures.
In V. Malyshkin (ed.), Parallel Computing Technologies: Proceedings of the 3th International Conference, Lect. Notes in Comp. Sci., Vol. 964, Springer, 1995, pp. 320-325

ISSN 0302-9743
ISBN 3-540-60222-4

Search LNCS articles for:
Etiemble, D. ; Germain, C.



Scanned first page:

Designed by the Supercomputer Software Department RAS, 1998