Cover -- Title Page -- Copyright -- Contents -- List of Tables -- List of Figures -- List of Listings -- Preface -- 1: Performance Evaluation -- 1.1. Performance evaluation -- 1.2. Performance versus resources provisioning -- 1.2.1. Performance indicators -- 1.2.2. Resources provisioning -- 1.3. Methods of performance evaluation -- 1.3.1. Direct study -- 1.3.2. Modeling -- 1.4. Modeling -- 1.4.1. Shortcomings -- 1.4.2. Advantages -- 1.4.3. Cost of modeling -- 1.5. Types of modeling -- 1.6. Analytical modeling versus simulation -- PART 1: Simulation -- 2: Introduction to Simulation -- 2.1. Presentation -- 2.2. Principle of discrete event simulation -- 2.2.1. Evolution of a event-driven system -- 2.2.2. Model programming -- 2.2.2.1. Scheduler -- 2.2.2.2. Object-oriented programming -- 2.3. Relationship with mathematical modeling -- 3: Modeling of Stochastic Behaviors -- 3.1. Introduction -- 3.2. Identification of stochastic behavior -- 3.3. Generation of random variables -- 3.4. Generation of U(0, 1) r.v. -- 3.4.1. Importance of U(0, 1) r.v. -- 3.4.2. Von Neumann's generator -- 3.4.3. The LCG generators -- 3.4.3.1. Presentation -- 3.4.3.2. Properties -- 3.4.3.2.1. MLCG with M = 2k -- 3.4.3.2.2. MLCG with M primer number -- 3.4.3.3. Examples of LCG -- 3.4.4. Advanced generators -- 3.4.4.1. Principle -- 3.4.4.2. Mersenne Twister generator -- 3.4.4.3. L'Ecuyer's generator -- 3.4.5. Precaution and practice -- 3.4.5.1. Nature of the PRNG -- 3.4.5.2. Choice of seed -- 3.4.5.3. Multiples substreams of RNG -- 3.4.5.3.1. Principle -- 3.4.5.3.2. Example of OMNeT++ -- 3.4.5.4. Quality of the PRNG -- 3.5. Generation of a given distribution -- 3.5.1. Inverse transformation method -- 3.5.2. Acceptance-rejection method -- 3.5.2.1. Case of finite support -- 3.5.2.2. Generalized version -- 3.5.3. Generation of discrete r.v.

3.5.3.1. Case of the finite discrete r.v. -- 3.5.3.2. Case of countably infinite discrete r.v. -- 3.5.4. Particular case -- 3.5.4.1. Composition -- 3.5.4.2. Convolution -- 3.6. Some commonly used distributions and their generation -- 3.6.1. Uniform distribution -- 3.6.1.1. Utilization -- 3.6.1.2. Parameters -- 3.6.1.3. Generation -- 3.6.2. Triangular distribution -- 3.6.2.1. Utilization -- 3.6.2.2. Parameters -- 3.6.2.3. Generation -- 3.6.3. Exponential distribution -- 3.6.3.1. Utilization -- 3.6.3.1.1. Arrival process -- 3.6.3.1.2. Memoryless phenomena -- 3.6.3.2. Parameter -- 3.6.3.3. Generation -- 3.6.4. Pareto distribution -- 3.6.4.1. Utilization -- 3.6.4.2. Parameters -- 3.6.4.3. Generation -- 3.6.5. Normal distribution -- 3.6.5.1. Utilization -- 3.6.5.2. Parameters -- 3.6.5.3. Generation -- 3.6.6. Log-normal distribution -- 3.6.6.1. Utilization -- 3.6.6.2. Parameters -- 3.6.6.3. Generation -- 3.6.7. Bernoulli distribution -- 3.6.7.1. Utilization -- 3.6.7.2. Parameter -- 3.6.7.3. Generation -- 3.6.8. Binomial distribution -- 3.6.8.1. Utilization -- 3.6.8.2. Parameters -- 3.6.8.3. Generation -- 3.6.9. Geometric distribution -- 3.6.9.1. Utilization -- 3.6.9.2. Parameter -- 3.6.9.3. Generation -- 3.6.10. Poisson distribution -- 3.6.10.1. Utilization -- 3.6.10.2. Parameter -- 3.6.10.3. Generation -- 3.7. Applications to computer networks -- 4: Simulation Languages -- 4.1. Simulation languages -- 4.1.1. Presentation -- 4.1.2. Main programming features -- 4.1.3. Choice of a simulation language -- 4.2. Scheduler -- 4.3. Generators of random variables -- 4.4. Data collection and statistics -- 4.5. Object-oriented programming -- 4.6. Description language and control language -- 4.7. Validation -- 4.7.1. Generality -- 4.7.2. Verification of predictions -- 4.7.3. Some specific and typical errors -- 4.7.4. Various tests.

4.7.4.1. Parameter setting and continuity -- 4.7.4.2. Robustness -- 5: Simulation Running and Data Analysis -- 5.1. Introduction -- 5.2. Outputs of a simulation -- 5.2.1. Nature of the data produced by a simulation -- 5.2.2. Stationarity -- 5.2.3. Example -- 5.2.4. Transient period -- 5.2.5. Duration of a simulation -- 5.3. Mean value estimation -- 5.3.1. Mean value of discrete variables -- 5.3.2. Mean value of continuous variables -- 5.3.3. Estimation of a proportion -- 5.3.4. Confidence interval -- 5.4. Running simulations -- 5.4.1. Replication method -- 5.4.2. Batch-means method -- 5.4.3. Regenerative method -- 5.5. Variance reduction -- 5.5.1. Common random numbers -- 5.5.2. Antithetic variates -- 5.6. Conclusion -- 6: OMNeT++ -- 6.1. A summary presentation -- 6.2. Installation -- 6.2.1. Preparation -- 6.2.2. Installation -- 6.3. Architecture of OMNeT++ -- 6.3.1. Simple module -- 6.3.2. Channel -- 6.3.3. Compound module -- 6.3.4. Simulation model (network) -- 6.4. The NED langage -- 6.5. The IDE of OMNeT++ -- 6.6. The project -- 6.6.1. Workspace and projects -- 6.6.2. Creation of a project -- 6.6.3. Opening and closing of a project -- 6.6.4. Import of a project -- 6.7. A first example -- 6.7.1. Creation of the modules -- 6.7.1.1. The module MySource -- 6.7.1.1.1. Description -- 6.7.1.1.2. Implementation -- 6.7.1.2. The module MySink -- 6.7.1.3. The simulation model -- 6.7.2. Compilation -- 6.7.3. Initialization -- 6.7.4. Launching of the simulation -- 6.8. Data collection and statistics -- 6.8.1. The Signal mechanism -- 6.8.2. The collectors -- 6.8.3. Extension of the model with statistics -- 6.8.3.1. Put probes in MySink.cc -- 6.8.3.2. Configuration through MySink.ned -- 6.8.4. Data analysis -- 6.9. A FIFO queue -- 6.9.1. Construction of the queue -- 6.9.1.1. Description -- 6.9.1.2. Insertion of a message -- 6.9.1.3. Service.

6.9.1.4. Sending of a message -- 6.9.1.5. Statistics -- 6.9.1.6. The FIFO queue model -- 6.9.2. Extension of MySource -- 6.9.3. Configuration -- 6.10. An elementary distributed system -- 6.10.1. Presentation -- 6.10.2. Coding -- 6.10.2.1. Namespace -- 6.10.2.2. Specific message format -- 6.10.2.3. Coding of the TRNode module -- 6.10.3. Modular construction of a larger system -- 6.10.4. The system -- 6.10.5. Configuration of the simulation and its scenarios -- 6.11. Building large systems: an example with INET -- 6.11.1. The system -- 6.11.2. Ethernet card with LLC -- 6.11.3. The new entity MyApp -- 6.11.3.1. Description of MyApp -- 6.11.3.2. Implementation of MyApp -- 6.11.4. Simulation -- 6.11.5. Conclusion -- PART 2: Queueing Theory -- 7: Introduction to the Queueing Theory -- 7.1. Presentation -- 7.2. Modeling of the computer networks -- 7.3. Description of a queue -- 7.4. Main parameters -- 7.5. Performance indicators -- 7.5.1. Usual parameters -- 7.5.2. Performance in steady state -- 7.6. The Little's law -- 7.6.1. Presentation -- 7.6.2. Applications -- 8: Poisson Process -- 8.1. Definition -- 8.1.1. Definition -- 8.1.2. Distribution of a Poisson process -- 8.2. Interarrival interval -- 8.2.1. Definition -- 8.2.2. Distribution of the interarrival interval Δ -- 8.2.3. Relation between N(t) and Δ -- 8.3. Erlang distribution -- 8.4. Superposition of independent Poisson processes -- 8.5. Decomposition of a Poisson process -- 8.6. Distribution of arrival instants over a given interval -- 8.7. The PASTA property -- 9: Markov Queueing Systems -- 9.1. Birth-and-death process -- 9.1.1. Definition -- 9.1.2. Differential equations -- 9.1.3. Steady-state solution -- 9.2. The M/M/1 queues -- 9.3. The M/M/∞ queues -- 9.4. The M/M/m queues -- 9.5. The M/M/1/K queues -- 9.6. The M/M/m/m queues -- 9.7. Examples.

9.7.1. Two identical servers with different activation thresholds -- 9.7.2. A cybercafe -- 10: The M/G/1 Queues -- 10.1. Introduction -- 10.2. Embedded Markov chain -- 10.3. Length of the queue -- 10.3.1. Number of arrivals during a service period -- 10.3.2. Pollaczek-Khinchin formula -- 10.3.3. Examples -- 10.4. Sojourn time -- 10.5. Busy period -- 10.6. Pollaczek-Khinchin mean value formula -- 10.7. M/G/1 queue with server vacation -- 10.8. Priority queueing systems -- 11: Queueing Networks -- 11.1. Generality -- 11.2. Jackson network -- 11.3. Closed network -- PART 3: Probability and Statistics -- 12: An Introduction to the Theory of Probability -- 12.1. Axiomatic base -- 12.1.1. Introduction -- 12.1.1.1. Sample space -- 12.1.1.2. Events -- 12.1.1.3. Probability -- 12.1.2. Probability space -- 12.1.2.1. σ-Algebra -- 12.1.2.2. Probability measure -- 12.1.2.3. Basic properties -- 12.1.3. Set language versus probability language -- 12.2. Conditional probability -- 12.2.1. Definition -- 12.2.2. Law of total probability -- 12.3. Independence -- 12.4. Random variables -- 12.4.1. Definition -- 12.4.2. Cumulative distribution function -- 12.4.3. Discrete random variables -- 12.4.3.1. Definition -- 12.4.3.2. Mathematical expectation -- 12.4.3.3. Probability generating function -- 12.4.4. Continuous random variables -- 12.4.4.1. Definition -- 12.4.4.2. Mathematical expectation -- 12.4.4.3. The Laplace transform -- 12.4.5. Characteristic function -- 12.5. Some common distributions -- 12.5.1. Bernoulli distribution -- 12.5.2. Binomial distribution -- 12.5.3. Poisson distribution -- 12.5.4. Geometric distribution -- 12.5.5. Uniform distribution -- 12.5.6. Triangular distribution -- 12.5.7. Exponential distribution -- 12.5.8. Normal distribution -- 12.5.9. Log-normal distribution -- 12.5.10. Pareto distribution.

12.6. Joint probability distribution of multiple random variables.