HiPC International Conference On High Performance Computing
About HiPC Steering Committee Sponsors Archive Home HiPC 2002 Home
         
     
HiPC 2001 - Hyderabad, India - December 17-20
Charminar - click for pdf
Advance Program
Workshops
Tutorials
Poster/Presentation Session
Program Highlights
Information for India-Based Participants
Travel and Local Information
Accomidation
Sight Seeing Tour

Conference Organization

Author Resources
Call for Papers
Registration
 
 

Tutorials

2 :0 0 p m - 6 :0 0 p m
TUTORIAL VII
High Performance CORBA

Shikharesh Majumdar
Carleton University

Audience: This tutorial is meant for researchers, system designers, users of CORBA-based sys-tems and products, and graduate students. Some background in distributed processing is required. A minor background in CORBA is desirable but not essential.

Course Description: Heterogeneity is natural in systems built using the popular Distributed Object Computing (DOC) paradigm. Different system components in such systems often use different programming languages and operating systems. Middleware provides inter-operability between clients and servers in such heterogeneous DOC systems. Common Object Request Broker Architecture (CORBA) is a middleware standard proposed by the Object Management Group (OMG). Most of the available commercial CORBA compliant middleware products provide inter-operability, but often incur a performance penalty. High scalability and low latency are crucial in many applications that include telecommunication products, process control systems, and other performance demanding applications. This tutorial will address these performance issues in CORBA-based systems. Three different classes of performance optimization techniques, guidelines for application design, adopting an appropriate client-middleware-server interaction architecture and techniques that exploit limited heterogeneity in systems will be described. Sample experimental results demonstrating their effectiveness will be discussed.

Lecturer: Shikharesh Majumdar is a Professor at the Dept. of Systems and Computer Engg. at Carleton University in Ottawa. He holds an M. Sc. and a Ph. D. degree in Computational Science from University of Saskatchewan, Saskatoon. Dr. Majumdar's research interests are in the areas of parallel and distributed processing, operating systems, and performance evaluation. He has authored over fifty technical papers, and has won a best presentation and a best paper award in two international IEEE conferences. He is the Associate Editor of the IEEE TC on Operating Systems Bulletin, and has served in the program and organizing committees of various international conferences. He is a member of the ACM and a Distinguished Visitor for the IEEE Computer Society.