Navigation

13S114PRS - Computer Systems Performance

Course specification
Course title Computer Systems Performance
Acronym 13S114PRS
Study programme Software Engineering
Module
Type of study bachelor academic studies
Lecturer (for classes)
Lecturer/Associate (for practice)
Lecturer/Associate (for OTC)
ESPB 6.0 Status mandatory
Condition None
The goal Teaching of the basic concepts of computer performance analysis and its application areas. Explanation of simplified models of computer system components (processors, memory, disks). Providing students with the ability to model and analyze computer systems and networks using stochastic methods and the mean value analysis (MVA) applied to open and closed queuing networks.
The outcome Students will be able to: select the appropriate modeling technique, depending on the characteristics of the computer component or system, set up an abstract model, specify the assumptions and approximations for the specific model, calculate performance indicators based on a set of given parameters, critically discuss the obtained numerical performance indicators, and interpret the results.
Contents
Contents of lectures Techniques and applications of computer performance analysis. Performance of the processors, memory subsystem and disks. Modeling of computer systems based on queuing networks. Poisson process. Birth and death formula. Exponential and non-exponential models. Pollaczek–Khinchine formula. Closed networks. Open networks. Central server networks. Stochastic and MVA analysis of systems and networks.
Contents of exercises Problems and examples that follow the order of topics presented during lectures: disk modeling based on linear, discrete and continuous models, examples of the system modeling based on single server queue, cyclic model of multiprogramming, central server network, Buzen's algorithm. Interactive systems and their analysis using stochastic methods and the MVA algorithm
Literature
  1. Rai Jain, "The Art of Computer Systems Performance Analysis," John Wiley and Sons, 1991.
  2. William J. Stewart, "Probability, Markov Chains, Queues, and Simulation: The Mathematical Basis of Performance Modeling," Princeton University Press, 2009. (Original title)
  3. Daniel A. Menasce, Lawrence W. Dowdy, Virgilio A.F. Almeida,"Performance by Design: Computer Capacity Planning By Example," Prentice Hall, 2004.
  4. Computer Systems Performance, presentations used on lectures and auditory exercises, Web site of the Department of Computer Engineering and Informatics, School of Electrical Engineering, http://rti.etf.bg.ac.rs/rti/prs/materijali/
Number of hours per week during the semester/trimester/year
Lectures Exercises OTC Study and Research Other classes
2 2 1
Methods of teaching Lectures and auditory exercises, individual work of students on assignments and projects.
Knowledge score (maximum points 100)
Pre obligations Points Final exam Points
Activites during lectures 0 Test paper 70
Practical lessons 0 Oral examination 0
Projects
Colloquia 0
Seminars 30