Parallel computers and parallel algorithms
125. R. P. Brent, Parallel computers and parallel algorithms,
Australian Computer Science Communications
13, 1 (1991), 1 (abstract of invited paper).
Abstract:
dvi (3K),
pdf (30K),
ps (28K).
Abstract
The era of serial computation, typified by Von Neumann computers and Turing
machines, is coming to an end. Parallel computation is now central
to Computer Science, and its importance will increase in the future.
I shall discuss the motivation for parallel computation (physical constraints
on serial computation, biological examples of parallel computation).
I shall describe some practical parallel computer architectures - tightly
coupled synchronous machines, loosely coupled asychronous networks, shared
memory and local memory machines, hypercubes and systolic arrays.
Concepts such as the
speedup and efficiency of parallel algorithms will be
defined and illustrated by some examples.
Comments
The invited paper was based on a George and Sandra
Forsythe memorial lecture presented at the
Department of Computer Science, Stanford University, February 1990.
For related work see
[128,
130,
131,
136].
Go to next publication
Return to Richard Brent's index page