Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site well.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!ucbvax!ucdavis!lll-crg!well!rchrd
From: rchrd@well.UUCP (Richard Friedman)
Newsgroups: net.general
Subject: Supercomputing Seminars in Nov. & Dec. from PSR
Message-ID: <210@well.UUCP>
Date: Thu, 3-Oct-85 17:40:37 EDT
Article-I.D.: well.210
Posted: Thu Oct  3 17:40:37 1985
Date-Received: Sat, 5-Oct-85 03:16:22 EDT
Organization: Whole Earth Lectronic Link, Sausalito, CA
Lines: 56
Keywords: Supercomputers Seminars Vector/Parallel-Processing CRAY PSR


PACIFIC-SIERRA RESEARCH announces two seminars on supercomputers
to be given in Northern California later this year:

Nov. 19-22 1985:  "Efficient FORTRAN Techniques for Vector Processors"
Dec.  3- 6 1985:  "Efficient FORTRAN for the CRAY-1,-2, and X-MP"

Both seminars will be held at the Placerville, California offices of
PSR (Placerville is about 45 minutes East of Sacramento on Highway 50).

Seminar Descriptions
--------------------

"Efficient FORTRAN Techniques for Vector Processors"

This seminar introduces the current state of the art of vector and parallel
processing.  A variety of machine architectures are presented, including
the CRAY -1, -2, and X-MP, CDC Cyber 205, Fujitsu VP-100, Convex C1,
Alliant FX/8, and others.

The FORTRAN environments of each system are analyzed from the point of view
of scalar and/or vector optimization strategies.  Using live examples of
FORTRAN code, we discuss the optimization techniques for improving scalar,
vector, and I/O performance for each system, and compare the results.
While many of the optimization techniques are similar from machine to
machine, there are some techniques which should be used only on certain
mainframes.  These issues are discussed in detail with real examples.

"Efficient FORTRAN Programming on the CRAYs"

This seminar teaches optimization and conversion techniques for the full
range of CRAY systems.	The course includes an introduction to vector
processing, a discussion of the architecture of the various CRAY systems,
and an in-depth look at the CRAY FORTRAN compiler CFT and its vectorization
abilities.  After this background information is presented, optimization
techniques for improving FORTRAN code performance, both scalar and vector,
are presented.	The seminar concludes with some I/O optimization guidelines
and a brief discussion of CRAY Job Control Language.

Both seminars include extensive student workbooks with interesting
assignment problems.

The instructors for both seminars are John Levesque and Joel Williamson.
Each instructor has over a decade's worth of experience optimizing
application codes for state-of-the-art supercomputers.

Cost: Each four-day seminar is $800 per attendee (includes workbook).

For further information, course, and room reservations, call Ms Tracey
Walsaw, PSR, 916 621 1600.
-- 
     
    [rchrd] = Richard Friedman
              Pacific-Sierra Research, 2855 Telegraph #415
              Berkeley, CA 94705 (415) 540 5216
    UUCP: {dual,hplabs,ptsfa,apple}!well!rchrd


Brought to you by Super Global Mega Corp .com