Tuesday, February 14, will be devoted to some hot topics in computer architecture. These topics include Symmetrical Multiprocessing, Massively Parallel Procesors, and Networks of Workstations. This week's hand-outs include five papers for you to read for the class:

Lucien M Censier and Paul Feautrier, "A New Solution to Coherence Problems in Multicache Systems", IEEE Transactions on Computers, 27(12):1112-1118, December 1978.

C. P. Thacker, L. C. Stewart, and E. H. Satterwaite Jr., "Firefly: a Multiprocessor Workstation", IEEE Transactions on Computers, 37(8):909-920, August 1988.

Anant Agarwal, Ricardo Bianchini, David Chaiken, Kirk Johnson, David Kranz, John Kubiatowicz, Beng-Hong Lim, Ken Mackenzie, and Donald Yeung, "The MIT Alewife Machine: Architecture and Performance". An abbreviated version of this will appear in ISCA '95.

Charles E. Leiserson, Zahi S. Abuhamdeh, David C. Douglas, Carl R. Feynman, Mahesh N. Ganmukhi, Jeffrey V. Hill, W. Daniel Hillis, Bradley C. Kuszmaul, Margaret A. St. Pierre, David S. Wells, Monica C. Wong, Shaw-Wen Yang, and Robert Zak, "The Network Architecture of the Connection Machine CM-5." An early version of this paper appeared in the 1992 ACM Symposium on Parallel Algorithms and Architectures, pp. 272-285, June 1992, San Diego, California.

Thomas E. Anderson, David E. Culler, and David A. Patterson, "A Case for NOW (Networks of Workstations)", to appear in IEEE Micro, 1995.

Assignment

Due next week is a one-page paper, focusing on the NOW paper and its relation to the other papers. Identify one or two ways in which the other papers support or contradict the NOW arguments. Do you agree with the arguments? How do you expect NOW to scale and evolve in the future?