Contents 1 Cloud Computing Resource Management and Scheduling Based on a Banking Model 1 1.1 Introduction 1 1.2 Inspiration for Computational Clouds 2 1.3 Banking Model Based Cloud Computing Resource Management and Scheduling 3 1.3.1 The Cloud Computing Requirements Analysis 4 1.3.2 The Technical Cloud Environment Requirement 5 1.4 Contributions to Research 6 1.4.1 Banking Model for Cloud Computing Resource Management 6 1.4.2 Optimal Deposit loan Ratio 7 1.4.3 Risk Mitigation and Prediction Management 7 1.4.4 Pricing Scheme of the Cloud Resource over the Lifecycle 8 1.4.5 Cloud Computing Resource Management and Distribution Based on the Pareto Equilibrium 9 1.4.6 Building a Real Cloud Computing Platform Based on Open Sourcing 10 1.5 Organization 11 References 12 2 Background Study on Cloud Computing, Resource Management and Scheduling 14 2.1 Computational Clouds 14 2.2 The Evolution of Cloud Computing Technology 15 2.2.1 Parallel Computation 15 2.2.2 Distributed Computing 17 2.2.3 The Main Difference Between Distributed Computing and Parallel Computing 17 2.2.4 Grid Computing 18 2.3 Cloud Computing 21 2.3.1 Background Research 21 2.3.2 The Definition of Cloud Computing 24 2.3.3 The Taxonomy of Cloud Computing 25 2.3.4 Internal Components of Cloud Computing 28 2.3.5 Main Differences Between Cloud Computing and Grid Computing 29 2.4 Cloud Resource Management 31 2.4.1 Main Strategies of Cloud Resource Management 31 2.4.2 The Taxonomy of Cloud Resource Management 32 2.4.3 Key Cloud Resource Management Technology 46 2.5 Summary 48 References 48 3 Economics and Cloud Computing Resource Management 52 3.1 Survey of Economic Theories Based on the Grid Resource Management Project 52 3.2 Economic Theories can Provide a Solution to Solve Cloud Computing Resource Management Issues 54 3.2.1 Cloud Computing as a Business driven Technology 54 3.2.2 Cloud Computing Technology Fit in with the Needs of the Society and Economic Laws 55 3.3 Using Economic Theories in Cloud Computing Resource Management 55 3.3.1 The Law of Demand and Supply in Cloud Computing Environment 55 3.3.2 The Law of Diminishing Marginal Returns in the Cloud Computing Environment 56 3.3.3 Monopolies in the Cloud Computing Environment 56 3.4 Summary 57 References 57 4 Problem Identification 59 4.1 Resource Accounting 59 4.2 Resource Scheduling 60 4.3 Cloud Computing Resource Transaction Risk Mitigation and Coping 61 4.4 The QoS Issue 62 References 62 5 Research Approaches to Banking Models for Cloud Computing Resource Management 63 5.1 Banking Model 64 5.2 How Does the Cloud Computing Follow the Real Bank to Do Transaction 66 5.2.1 Optimal Deposit loan Ratio Theory in Cloud Banks 68 5.2.2 Identifying Factors Affecting the Cloud Bank 69 5.3 The Pricing Schema for Cloud Computing 69 5.4 Avoiding Banking Risk in the Transaction 69 5.5 Cloud Bank Scheduling and the Pareto Optimality 70 5.6 Interior Components of the Cloud Bank 71 5.7 Summary 72 References 72 6 Research Approaches for Risk Mitigation and Coping 74 6.1 The Risk Mitigation and Management in Commercial Banks 74 6.2 The Risk Mitigation in the Cloud Bank 76 6.2.1 The Classification of Risks in Cloud Bank 77 6.2.2 The Strategy of Risk Mitigation 78 6.2.3 The Strategy of the Risk Coping 90 6.3 Experiment Setup 92 6.4 Analysis of Experimental Results 98 6.5 Summary 99 References 100 7 Research Approaches for the Pricing Scheme of the Cloud Bank in the Price Lifecycle 102 7.1 The Centralized Synchronous Algorithm 102 7.1.1 The Theory of Optimal Deposit loan Algorithm 103 7.1.2 The Resource Management Model Based on Optimal Deposit Loan Algorithm 103 7.1.3 Single Resource Pricing Underlying the Cloud Bank Model 106 7.1.4 About the Optimal Deposit loan Algorithm 107 7.2 Distributed Price Adjustment Algorithm 110 7.2.1 Pricing Scheme of Cloud Resources in the Initial Stage 110 7.2.2 Pricing Scheme of Cloud Resources in a Stable Stage 116 7.3 The Service Level Agreement (SLA) of the Cloud Bank Model 118 7.3.1 Why We Need CBSLA 119 7.3.2 The CBSLA Framework 120 7.3.3 Signature Process of the CBSLA Contract 121 7.3.4 Generation of the CBSLA 124 7.4 Summary 124 References 125 8 Research Approaches for the Pareto Optimality Based Scheduling 126 8.1 The Concept of Pareto Optimality 126 8.1.1 Pareto Optimality 127 8.1.2 Pareto Improvement 127 8.2 Cloud Banks Achieve Optimal Resources Allocation by Pareto Theory 130 8.3 The Extended Pareto Optimality Model 131 8.3.1 Relative Proof 132 8.3.2 To Solve the Problem Under M×N, Pareto Optimality 133 8.4 Cloud Banks Achieve Optimal Resources Allocation by Pareto Optimality Theory 135 8.5 Improvement of PO based Allocation Strategy 136 8.6 The Steps of Dynamic Simulation 138 8.7 The Simulation Environment Set up 139 8.8 Running the CloudSim Instance 141 8.9 Analysis of Experimental Results 144 8.1 0Summary 147 References 148 9 The Real Laboratory Platform: IaaS Based Cloud Computing Platform 150 9.1 Introduction 150 9.2 Setting up the IaaS Based Cloud Computing Environment 151 9.2.1 The Comparison of the Two Kinds of the Platform Structures 151 9.2.2 Introduction of Eucalyptus 152 9.2.3 EUCALYPTUS Platform Advantage 153 9.2.4 EUCALYPTUS Platform Framework 154 9.2.5 EUCALYPTUS Components 155 9.2.6 EUCALYPTUS Configuration 156 9.2.7 EUCALYPTUS Installation Readiness 158 9.2.8 Installation of EUCALYPTUS Technical Route 159 9.2.9 Specific Methods of EUCALYPTUS Installation 160 9.3 Cloud Computing Simulator:CloudSim in Use 161 9.4 The Structure of the CloudSim 161 9.5 Summary 163 References 164 10 Conclusions and Future Directions 165 10.1 Summary 165 10.2 Conclusions 166 10.3 Future Directions 168 10.3.1 Supporting Accounting and Visualization 168 10.3.2 Supporting Complex Service and Task Description 168 10.3.3 Supporting Real Cloud Computing Environment Experiment Platform 169 10.3.4 Supporting a Variety of Risks in a Cloud Computing Environment 169