Synopsis

The objective of this course is to provide students a working knowledge of computer performance evaluation and capacity planning.  It covers the fundamental techniques in performance evaluation of computer systems including measurement and mathematical modeling. The module is divided into four main parts: introduction to performance analysis and capacity planning, measurement techniques and tools, analytic modeling techniques and case studies. Topics include: performance analysis overview; measurement techniques and tools covering performance metrics, workload characterization, instrumentation; analytical modeling techniques covering operational analysis, stochastic queuing network analysis, performance of client-server architectures, principles of scalable performance.

 

Instructor

            Assoc Professor Teo Yong Meng

          Room  : S14, Level 6, #06-12

          E-mail  : teoym AT comp.nus.edu.sg

          URL     : www.comp.nus.edu.sg/~teoym
 

Lecture                      : Wed, 1830-2030, COM1/204

Consultation Hours : Wed, 0900-1100

Examination             : 6 May 2009, evening (to be confirmed)

Modular Credits       : 4

 

Main Textbooks

·      The Art of Computer Systems Performance Analysis: Techniques for Experimental Design, Measurement, Simulation and Modeling, R. Jain, John-Wiley, 1991.   [Errata]

·        Quantitative System Performance, E.D. Lazowska et al., Prentice-Hall, 1984.

·        Measuring Computer Performance - A Practitioner's Guide, D.J. Lilja, Cambridge University Press, 2000.

 

Module Assessment:

·        continuous assessment - 60%  

·        final examination - 40% (open book exam)


This document, cs5239.htm, has been accessed 3506 times since 14-Jan-09 15:37:48 SGT. This is the 2nd time it has been accessed today.

A total of 1111 different hosts have accessed this document in the last 1123 days; your host, 38.107.179.220, has accessed it 1 times.

If you're interested, complete statistics for this document are also available, including breakdowns by top-level domain, host name, and date.