By Topic

Comments on "A Massive Memory Machine"

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
$33 $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

2 Author(s)
J. R. Goodman ; Department of computer Science, University of Wisconsin ; H. C. Young

Garcia-Molina, Lipton, and Valdes [1] introduced a new machine architecture called "massive memory machines" (MMM). The primary application of their proposed architecture was for so-called memory bound computations. In this correspondence we argue: 1) that massive memories will likely become feasible, but will be most effective with much more powerful processors, and 2) that a massive memory on the proposed machine will perform poorly in the same cases that virtual memory performs poorly: whenever there is poor locality of memory reference. Other problems with the architecture are also discussed. These related issues include: 1) the infeasibility of large on-chip dual port memory, 2) the support of multiprocessing on an ESP, 3) the possibility of memory prerequest, 4) the potential of trading program size for execution time, and 5) the time required for clearing the entire memory.

Published in:

IEEE Transactions on Computers  (Volume:C-35 ,  Issue: 10 )