By Topic

Modeling parallel communication

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Karp, Richard M. ; California Univ., Berkeley, CA, USA

The LogP model provides a framework for analyzing the performance of algorithms on distributed-memory multiprocessors. The model takes into account both the cost of computation and the cost of communication. It describes a multiprocessor in terms of four parameters, representing computation power, communication bandwidth, communication latency and the degree of overlap between computation and communication. In this paper we discuss the rationale for the different features of the LogP model. We then compare the model with other models having similar aims, such as network-based models, Valiant's BSP model, and the PRAM(m) model. Finally, we call upon the work of a number of students at Berkeley to illustrate the application of the model to problems of broadcasting, sorting, FFT computation, summing, prefix summing and solution of triangular and tridiagonal systems

Published in:

Parallel Processing Symposium, 1995. Proceedings., 9th International

Date of Conference:

25-28 Apr 1995