Cover -- Title Page -- Copyright Page -- Contents -- Preface -- PART I VERIFICATION -- 1 Modeling and Verification of Distributed Systems Using Markov Decision Processes -- 1.1 Introduction -- 1.2 Markov Decision Processes -- 1.3 Markov Decision Well-Formed Net formalism -- 1.4 Case study: Peer-to-Peer Botnets -- 1.5 Conclusion -- Appendices: Well-formed Net Formalism -- A.0.1 Syntax of Basic Predicates -- A.0.2 Markings and Enabling -- References -- 2 Quantitative Analysis of Distributed Systems in Stoklaim: A Tutorial -- 2.1 Introduction -- 2.2 StoKlaim: Stochastic Klaim -- 2.2.1 Klaim in a Nutshell -- 2.2.2 Syntactic Categories -- 2.2.3 StoKlaim Syntax -- 2.2.4 StoKlaim at Work -- 2.3 StoKlaim Operational Semantics -- 2.3.1 Rate Transition Systems -- 2.3.2 StoKlaim: RTS-based Semantics -- 2.4 MoSL: Mobile Stochastic Logic -- 2.5 jSAM: Java Stochastic Model-Checker -- 2.6 Leader Election in StoKlaim -- 2.6.1 As far as it can -- 2.6.2 Asynchronous Leader Election -- 2.7 Concluding Remarks -- References -- 3 Stochastic Path Properties of Distributed Systems: the CSLTA Approach -- 3.1 Introduction -- 3.2 The Reference Formalisms for System Definition -- 3.3 The Formalism for Path Property Definition: CSLTA -- 3.4 CSLTA at work: a Fault-Tolerant Node -- 3.5 Literature Comparison -- 3.6 Summary and Final Remarks -- References -- PART II EVALUATION -- 4 Failure Propagation in Load-Sharing Complex Systems -- 4.1 Introduction -- 4.2 Building Blocks -- 4.2.1 Coarse-grained Modeling -- 4.2.2 Abstract Mechanisms Impacting the Failure Occurrence -- 4.2.3 Parametric Distributions Revisited -- 4.2.4 Exponential Distribution -- 4.2.5 Weibull Distribution -- 4.2.6 Lognormal Distribution -- 4.2.7 Other Distributions -- 4.3 Sand Box for Distributed Failures -- 4.3.1 Failure Modes -- 4.3.2 LOS and Stress Rupture -- 4.4 Summary -- References.

5 Approximating Distributions and Transient Probabilities by Matrix Exponential Distributions and Functions -- 5.1 Introduction -- 5.2 Phase Type and Matrix Exponential Distributions -- 5.3 Bernstein Polynomials and Expolynomials -- 5.4 Application of BEs to Distribution Fitting -- 5.5 Application of BEs to Transient Probabilities -- 5.6 Conclusions -- References -- 6 Worst-Case Analysis of Tandem Queueing Systems Using Network Calculus -- 6.1 Introduction -- 6.2 Basic Network Calculus Modeling: Per-fl ow Scheduling -- 6.2.1 Service Curve -- 6.2.2 Arrival Curve -- 6.2.3 Delay and Backlog Bounds -- 6.2.4 Numerical Examples -- 6.3 Advanced Network Calculus Modeling: Aggregate Multiplexing -- 6.3.1 Aggregate-multiplexing Schemes -- 6.4 Tandem Systems Traversed by Several Flows -- 6.4.1 Model -- 6.4.2 Loss of the Tightness -- 6.4.3 Separated-flow Analysis -- 6.5 Mathematical Programming Approach -- 6.5.1 Blind Multiplexing -- 6.5.2 FIFO Multiplexing -- 6.6 Related Work -- 6.7 Numerical Results -- 6.8 Conclusions -- References -- 7 Cloud Evaluation: Benchmarking and Monitoring -- 7.1 Introduction -- 7.2 Benchmarking -- 7.2.1 Benchamrking State of Art -- 7.2.2 Benchmarking Big Data Services -- 7.3 Benchmarking with mOSAIC -- 7.4 Monitoring -- 7.4.1 Monitoring Problem Scenarios -- 7.4.2 Monitoring Problem Analysis -- 7.4.3 Monitoring State of the Art -- 7.5 Cloud Monitoring in mOSAIC's Cloud Agency -- 7.6 Conclusions -- References -- 8 Multiformalism and Multisolution Strategies for Systems Performance -- 8.1 Introduction -- 8.2 Multiformalism and Multisolution -- 8.3 Choosing the Right Strategy -- 8.4 Learning by the Experience -- 8.4.1 Distributed Transaction Processing -- 8.4.2 Service Oriented Architectures -- 8.4.3 Supervision of Distributed Information Systems -- 8.4.4 Big Data Architectures -- 8.4.5 Degradation for Software Aging.

8.4.6 Product Forms Exploitation -- 8.5 Conclusions and Perspectives -- References -- PART III OPTIMIZATION AND SUSTAINABILITY -- 9 Quantitative Assessment of Distributed Networks Th rough Hybrid Stochastic Modeling -- 9.1 Introduction -- 9.2 Modeling of Complex Systems -- 9.2.1 Classical Non State-space Models -- 9.2.2 State-space Models -- 9.2.3 High Level Formalisms -- 9.2.4 Stochastic Activity Networks -- 9.2.5 Adaptive Transition Systems -- 9.2.6 Analytical Solution vs Simulation -- 9.3 Performance Evaluation of KNXnet/IP Networks Flow Control Mechanism -- 9.3.1 Overview of KNX and KNXnet/IP -- 9.3.2 The KNXnet/IP Flow Control Mechanism -- 9.3.3 Modeling Hypotheses and Motivation for Using the SAN Formalism -- 9.3.4 KNX TP1 Communication Device Model -- 9.3.5 KNXnet/IP Router Model -- 9.3.6 Results -- 9.3.7 Model Settings -- 9.3.8 Analysis of Information Flow from Subnet1 to Subnetb -- 9.4 LCII: On-line Risk Estimation of A Power-Telco Network -- 9.4.1 Power Network -- 9.4.2 Stochastic model of the PN -- 9.4.3 Simulation of the Power Network -- 9.4.4 TELCO sites and backup batteries -- 9.4.5 Stochastic model of the batteries -- 9.4.6 The online Risk Estimator -- 9.5 Conclusion -- References -- 10 Design of IT Infrastructures of Data Centers: An Approach Based on Business and Technical Metrics -- 10.1 Introduction -- 10.2 Fundamental Concepts -- 10.2.1 Dependability -- 10.2.2 Reliability Importance -- 10.2.3 Factorial Experimental Design -- 10.2.4 Hierarchical Clustering -- 10.3 Business-Oriented Models -- 10.3.1 Infrastructure Cost -- 10.3.2 Infrastructure Revenue -- 10.3.3 Penalty -- 10.3.4 Profit -- 10.3.5 Additional Profit per Monetary Unit -- 10.4 Data Center Infrastructure Models -- 10.4.1 Modeling Strategy -- 10.4.2 Dependability Models -- 10.5 Methodology -- 10.5.1 Phase I: Problem Analysis -- 10.5.2 Phase II: System Modeling.

10.5.3 Phase III: Design Selection -- 10.6 Case Study - Data Center Design -- 10.6.1 Base Architectures -- 10.6.2 Modeling and Evaluation -- 10.7 Conclusion -- References -- 11 Software Rejuvenation and its Application in Distributed Systems -- 11.1 Introduction -- 11.2 Software rejuvenation scheduling classification -- 11.3 Software rejuvenation granularity classification -- 11.3.1 Physical node granularity rejuvenation -- 11.3.2 Operating system granularity rejuvenation -- 11.3.3 Virtual machine monitor/hypervisor rejuvenation granularity -- 11.3.4 Virtual machine rejuvenation granularity -- 11.3.5 Application rejuvenation granularity -- 11.3.6 Application component rejuvenation granularity -- 11.4 Methods, policies and metrics of soft ware rejuvenation -- 11.5 Software rejuvenation in distributed systems -- 11.6 Summary -- References -- 12 Machine Learning Based Dynamic Reconfiguration of Distributed Data Management Systems -- 12.1 Introduction -- 12.2 Methodologies -- 12.2.1 ML Approaches -- 12.3 Brief overview of Neural Networks -- 12.4 System Architecture and Performance Prediction Scheme -- 12.4.1 Model of the Data Grid Platform -- 12.4.2 Objective Functions -- 12.4.3 Platform Reconfiguration -- 12.5 Experimentation -- 12.5.1 Infinispan Overview -- 12.5.2 Experimental Settings -- 12.5.3 Results -- 12.6 Conclusions -- References -- 13 Going Green with the Networked Cloud: Methodologies and Assessment -- 13.1 Introduction -- 13.2 Modeling of Data Centre Power Consumption -- 13.2.1 CPU Power Dissipation -- 13.2.2 Server Power Consumption -- 13.2.3 Power Consumption in a Networked Environment -- 13.3 Energy Efficiency in the Cloud -- 13.3.1 Energy conservation techniques for servers -- 13.3.2 Power conservation techniques for networks -- 13.4 Performance Analysis Methodologies and Tools -- 13.4.1 Evaluation Metrics.

13.4.2 Performance Analysis Tools and Settings -- 13.5 Case Study: Performance Evaluation of Energy Aware Resource Allocation in the Cloud -- 13.5.1 Experimentation Setup -- 13.5.2 Numerical Results -- 13.6 Summary -- References -- Index -- EULA.