Performance Evaluation and Dimensioning of Networks and Systems

The main objectives of this course were to address the problem of resource management in computer networks and systems, introduce the main techniques of performance analysis and dimensioning (stochastic modelling, discrete event simulation and optimization) and their use in the context of networks and systems. During the course the following skills were aquired:

(i) Be able to evaluate the performance of networks and services; 
(ii) Be able to efficiently dimension networks and services using optimization techniques; 
(iii) Understand the performance/cost tradeoff in the resource management of networks and services; 
(iv) Be able to apply queuing theory in network performance analysis and understand its limitations; 
(v) Be able to properly handle network and service simulators. The course aims at training performance and planning engineers.

• The course of Performance Evaluation and Dimensioning of Networks and Systems was divided in three modules:

Module 1 - Discrete Event Simulation:

• The aim of the laboratory work is to study the techniques for generating random variables and stochastic processes, and for programming discrete event simulators and analyse its output data. Laboratory Report Nr1

Module 2 - Performance evaluation:

• The aim of this laboratory work is to study discrete-time Markov chains (DTMCs), continuous-time Markov chains (CTMCs), queuing systems, and their applications. Laboratory Report Nr2

Module 3 - Optimization:

• The aim of this laboratory work is to study the use of optimization methods in the management and dimensioning of networks and systems. Laboratory Report Nr3