Pre-requisites : NIL
Modelling philosophy and Principles: Discussion of several techniques of modelling; Introduction to Queuing theory and Markov chains: Queuing theory applied to study single and multiple server systems, statistical equilibrium, stationary processes, and ergodic processes; Monte Carlo simulation: Build simulation models from first principles, Discussion of the technique and theory of simulation, Start the simulation model as lab work; Mean Value analysis applied to multiprocessor systems; Workload Characterization: System dependent characterization, System independent characterization; Lagrangian and Eulerian views of job flow in systems; Communication Networks: Relation between bandwidth and latency; Sizing of Systems; Capacity Planning of systems; Software performance and Tuning; Software and Hardware monitors; Controlled experiments: Benchmarking and extrapolation of results to uncontrolled environment; Cloud Computing systems: performance framework.Laboratory Work: Simulation models will be explored using simulation tools.
1. Kai Hwang, Advanced Computer Architecture, McGraw Hill,Computer Science Series, 1993.
2. Krishna Kant, Introduction to Computer System Performance Evaluation, McGraw Hill International Science series, 1992.
3. Gunter Bolch, Stefan Greimer, Herman de Meer,and Kishore S. Trivedi, Queuing Networks and Markov Chains, John Wiley, Interscience, 1998.
4. Jean Walrand, Kallol Bagchi and George W. Zobrist, editors, Network Performance Modeling and Simulation, Gordon and Breach Science Publishers, 1998.
5. Domenico Ferrari, Computer Systems Performance Evaluation, Prentice Hall, 1978.
6. J. L. Hennessy and D. A. Patterson, Computer Architecture: A Quantitative Approach, Morgan Kaufmann, fourth edition, 2006.
7. Peter G. Harrison, Naresh M. Patel, Performance Modeling of Communication Networks and Computer Architectures, Addison Wesley Longman Publishing Co, 1992.