Richard A. Lethin

Last Edited 11/21/95

Recommended Books

Member of the Concurrent VLSI Architecture Group at the MIT Artificial Intelligence Lab . Research in computer architecture. Currently working on my PhD, studying heavy-load message-passing dynamics on the J-machine.

For the moment, I'm also assisting the compiler effort which will retarget the Multiflow Compiler to the M-machine.

Have worked as a designer on several hardware projects.

  • Multiflow TRACE 200 and 300 series VLIW Minisuper Computers. Designed the Floating Point Processor Board for the original 200 and the Janus upgrade based on the BIT chips that led to the 300 series.
  • MIT J-machine. Designed the RALU portion of the Message Driven Processor and wrote most of the simulators that we used to shake out bugs from the logic before fabbing it. Click here for a picture of the J-machine and me.

    • Some References

  • William J. Dally et. al (including me as one co-author), "The Message-Driven Processor: A Multicomputer Processing Node with Efficient Mechanisms", IEEE Micro, 1992.
  • William J. Dally et al (including me as one co-author) "The J-Machine: A Fine-Grain Concurrent Computer" Proceedings Information Processing 89.
  • Richard A. Lethin and William J. Dally, "MDP Design Tools and Methods", Proceedings of the International COnference on Computer Design: VLSI in Computers and Processors, 1992
  • Richard A. Lethin, "A Simulator for the Message-Driven Processor", Master's Thesis, MIT 1991.
  • Robert P. Colwell and Richard A. Lethin, "Latent Design Faults in the Development of Multiflow's TRACE/200", Accepted by IEEE Transactions on Reliability, expected pub date Fall '94.
  • William J. Dally, Scott D. Wills, and Richard Lethin, "Mechanisms for Parallel Computing", Proceedings of the NATO Advanced Study Institute on Parallel Computing and Distributed Memory Multiprocessors, Springer, 1991.
  • Richard A. Lethin, "Examination of Neural Darwinism," Massachusetts Institute of Technology Concurrent VLSI Architecture Group Memo # 44, 1992.
  • Richard A. Lethin, "A Random Diagnostic Generator for Superscalar Processors," Massachusetts Institute of Technology Concurrent VLSI Architecture Group Memo # 48, 1993.
  • Richard A. Lethin, "Parallel Computers and Commercial Applications," Massachusetts Institute of Technology Concurrent VLSI Architecture Group Memo # 49, 1992.
  • The Multiflow Trace is described in: Robert P. Colwell et al, "A VLIW Architecture for a Trace Scheduling Compiler", IEEE Transactions on Computers, August 1988.
  • I also wrote pixie for the Multiflow Trace, which allowed me to do most of the performance measurements reported in: P. G. Lowney et al, "The Multiflow Trace Scheduling Compiler," The Journal of Supercomputing, May, 1993 Volume 7, Number 1, pp. 51-152.

    • Interested in the current state of the theory of NP-hard optimization problems and approximation algorithms? Checkout this compendium by Crescenzi and Kann.

    My best friend.

    For fun, I once played Trombone in the Longwood Symphony Orchestra

    Here is a link to a friend who is a writer and poet.

    My arms are long. It's difficult for me to purchase shirts that fit. A friend in Hong Kong introduced me to a tailor named Ash Samtani who has made clothing for his family for years. His prices were so-so, but I ordered some shirts. They fit well. One batch of shirts I purchased had trouble with the buttons falling off. I'll probably order some more at some time. If you'd like you can contact his shop by sending email to samtani@hk.super.net. Ash's son, Luke, makes annual trips to the United States visiting most major cities and taking orders and measurements.

    I've found the eyeglass store called Opticus on Beacon Hill to be the best in Boston. They have more or less the same selection as everyone else (excluding shops like the one in Harvard Square that sells exclusive hand made titanium Italian frames for big $) but great service. Great service for me: they untwist my frames every time I go by, they never seem too busy to help, they keep track of their customers, and they're willing to do temporary repairs while new orders are being filled.

    The Red Rock Eater News Service is one of the few mailing lists to which I subscribe.

    This is Eugene Miya's parallel processing bibliography (old).

    I've set up a couple of mailing lists at the AI lab.

  • Electronic Cash Market (ECM) Trading Page.
  • The Boston Society for Digital Commerce.
  • Digerati.
  • The Melatonin Discussion List
    • Other misc.
  • Adjusting the office radiators in Tech Square
  • Fighting chronic post-nasal annoyance.
  • Poor service, rotten price, from a garage in Seattle.
  • Scoliosis mailing list.
  • Nellie. 
    • E-mail: lethin@ai.mit.edu
P-mail: 545 Technology Sq. Room 610
        Cambridge, MA 02139
Phone:  (617) 253-0972
Fax:    (617) 253-5060