Dataflow summer course 
Author Message
 Dataflow summer course

Dear Colleague,

Enclosed is an announcement for a 1-week course on Dataflow Architectures
and Languages that I will be teaching at MIT this summer.  It may be of
interest to you or to your colleagues.  Thank you.

There is a change in the course dates this year; please take note.

Arvind
Charles W. and Jennifer C. Johnson Professor
 of Computer Science and Engineering
MIT

----------------- COURSE ANNOUNCEMENT: PLEASE POST  ----------------

Parallel Computing: Dataflow Architectures and Languages
  (with Programming Laboratory on Monsoon Dataflow Machines)

Monday, June 15 through Friday, June 19, 1992

Massachusetts Institute of Technology
Summer Session Program 6.83s
                             ----------------
                              Course Abstract

The only thing holding back the widespread use of parallel computers
is software. Most of the difficulty of parallel programming is
attributable to our von Neumann legacy - imperative languages and
sequential processor architectures.  By switching to functional
languages, one may start writing parallel programs without even
realizing it.  Dataflow architectures further simplify the compilation
problem by providing cheap synchronization.

A central theme of the course is Id, an implicit parallel language.
The participant will get a chance to evaluate via laboratory
experience whether Id is a fad or a real alternative to standard
imperative languages extended for parallelism, such as C with threads,
Multi-Lisp, and fortran 9X.  The participant will also get an
opportunity to compare Id to purely functional languages.  Compilation
of Id for both dataflow and von Neumann machines will be discussed at
length.

The other theme of the course is dataflow architectures. We will
discuss why these architectures are better building blocks for
parallel computers than modern RISC architectures. Today's dataflow
architectures borrow much from traditional architectures; however,
they take the most aggressive approach to multi-threading, that is,
rapid context switching to tolerate long memory latencies and frequent
synchronization waits.  The participant will get hands on experience
on Monsoon dataflow machines produced by Motorola, and a chance to
conduct experiments on emulators of other dataflow machines.  We will
also discuss several supercomputer class dataflow machines that are
currently under construction.

                           ----------------
                           Course Outline:

Implicit Parallel Programming:
  Programming with higher-order functions and non-strict data
  structures; Rewrite rules and reduction; Algebraic and abstract data
  types; Arrays and I-structures; M-structures and non-determinism.

Architectures:
  Fundamental issues in high-performance parallel architectures;
  Static and dynamic dataflow machines; Split phase memory references;
  I-structure memory; Multi-threaded architectures; Hybrid von
  Neumann-dataflow architectures.

Compilation:
  Dataflow program graphs; Translation to dataflow graphs; Lambda-
  lifting and supercombinators; Loop, array and procedure call
  optimization.

Resource management and performance:
  Resource managers; Experimental results on MIT dataflow machines.

                           ----------------
Laboratory:
  Morning and afternoon lecture sessions will be followed by late-
  afternoon laboratory sessions in writing, debugging, running and
  analyzing the performance of Id programs on a Monsoon dataflow
  machine and on software emulators.  Experienced assistants will be
  available in the laboratory.
                           ----------------
The Target Audience:
  Understanding dataflow principles can benefit users and designers of
  all parallel systems, i.e., parallel languages, architectures,
  compilers and resource managers.  In addition to computer scientists
  and electrical engineers, the course is also useful for people
  working in scientific programming, signal processing, real-time
  computing and artificial intelligence.

Staff:
  The program will be taught by professor Arvind of the MIT Department
  of Electrical Engineering and Computer Science.

- ----------------
FOR MORE INFORMATION:

For a more detailed brochure (including application forms and
information about housing and fees), please contact:

    Ann Maderer
    Asst. to Prof. Arvind
    MIT Laboratory for Computer Science
    NE43-209
    545 Technology Square
    Cambridge, MA 02139, USA


    Tel: (617) 253-6837
    Fax: (617) 253-6652



Mon, 24 Oct 1994 06:43:05 GMT  
 
 [ 1 post ] 

 Relevant Pages 

1. Dataflow Course at MIT in August.

2. Dataflow Architectures and Languages Course Announcement

3. Course Announcement: Dataflow Architectures and Languages at MIT

4. REPOST: Dataflow Course at MIT in August

5. MIT Summer Course Announcement-July Version

6. MIT Summer Course Announcement

7. Summer Courses?

8. Summer Course on Intensive C++

9. MIT Summer Course Announcement-July Version

10. MIT Summer Course Announcement

11. MIT Summer Course Announcement-July Version

12. MIT Summer Course Announcement

 

 
Powered by phpBB® Forum Software