home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: git.cc.general,git.cc.talk,git.cc.faculty,git.cc.students,git.cc.students.undergrad,git.cc.students.grad,git.announce,git.general,git.events,atl.general
- Path: sparky!uunet!europa.asd.contel.com!darwin.sura.net!cc.gatech.edu!solaria.cc.gatech.edu!croft@cc.gatech.edu
- From: croft@cc.gatech.edu (Molly Croft)
- Subject: David Kuck here on Thursday, January 14
- Message-ID: <9301122323.AA24545@terminus.cc.gatech.edu>
- Sender: news@cc.gatech.edu
- Original-To: faculty@cc, faculty@ee, events@cc, mslist@cc, phdlist@cc
- Organization: Georgia Tech College of Computing
- Distribution: git
- Date: Tue, 12 Jan 1993 23:23:25 GMT
- Return-Path: <croft@cc.gatech.edu>
- Approved: git.cc.general,git.cc.talk,git.cc.faculty,git.cc.students,git.cc.students.undergrad,git.cc.students.grad,git.announce,git.general,git.events,atl.general@news.cc.gatech.edu
- X-Mailer: ELM [version 2.3 PL11]
- Lines: 60
-
-
- COLLEGE OF COMPUTING DISTINGUISHED LECTURE SERIES
-
- David J. Kuck, Professor
- Computer Science and Electrical and Computer Engineering
- Center for Supercomputing Research and Development
- University of Illinois at Urbana-Champaign
-
-
- "Cedar, Parallel Processing and the Greatest Grand Challenge"
-
-
- Thursday, January 14
- College of Computing, Room 17
- 3:30 p.m. -- Reception, 4:00 p.m. -- Lecture
-
-
- Abstract
-
- The key question about the future of supercomputing is not whether
- parallelism will be used but how it will be used. Parallelism is necessary
- because the increase in the fastest machine clock speeds over the past few
- decades is leveling off. After years of debate, the Federal High
- Performance Computing and Communications Initiative has now been launched.
- While HPCCI has focused much attention on parallel computing of Grand
- Challenge applications, it has also promised performance improvements that
- seem unachievable.
-
- The Greatest Grand Challenge (GGC) is to design parallel computing systems
- that can continue to produce speed increases into the indefinite future,
- through parallelism rather than clock speed advances. To deliver high
- parallel system performance, we need well-designed algorithms, system
- software, and compatible architectures. Much work is needed in performance
- evaluation and the use of performance results in system design. The talk
- will survey system performance issues and give real machine results using
- the Perfect codes as well as examples from the Cedar system.
-
- Bio
-
- David J. Kuck is a Professor of Computer Science and Electrical and Computer
- Engineering at the University of Illinois at Urbana-Champaign, where he has
- been a faculty member since 1965. From 1984 to 1992, he was Director of the
- Center for Supercomputing Research and Development, which he organized in
- 1984. He founded Kuck and Associates, Inc., Champaign, IL, in 1979 and is
- Chairman of the Board of Directors. He is currently engaged in the
- development and refinement of the Cedar parallel processing system and, in
- general, in theoretical and empirical studies of various machine and
- software organization problems, including parallel, pipeline and
- multiprocessor computation methods, interconnection networks, memory
- hierarchies, and the compilation of ordinary programs for parallel machines.
- Dr. Kuck received a B.S.E.E. in 1959 from the University of Michigan and a
- Ph.D from Northwestern University in 1963. He is a Fellow of IEEE and AAAS
- and a member of the National Academy of Engineering.
-
- --
-
- Molly Ford Croft
- croft@cc.gatech.edu
- College of Computing, Room 149
- (404) 853-2682
-