Large Scale Network-Centric Distributed Systems

18

Game-Based Models of Grid User's Decisions in Security-Aware Scheduling

Joanna Kolodziej, Samee U. Khan, Lizhe Wang, and Dan Chen

Contents

18.1 Introduction

18.2 Security-Aware Scheduling Problems in Computational Grids

18.2.1 Generic Model of Secure Grid Clusters

18.2.2 Security Criterion in Grid Scheduling

18.2.2.1 Risky Mode

18.2.3 Requirements of Grid End Users for Scheduling

18.3 Game Models in Security-Aware Grid Scheduling

18.3.1 Symmetric and Asymmetric Non-cooperative Games of the End Users

18.3.1.1 Non-cooperative Symmetric Game

18.3.1.2 Asymmetric Scenario—Stackelberg Game

18.3.2 Cooperative and Semi-cooperative Game Scenarios

18.3.3 Online Scheduling Games

18.4 Case Study: Approximating the Equilibrium States of the End Users' Symmetric Game Using the Genetic Metaheuristics

18.4.1 Specification of the Game

18.4.1.1 Characteristics of Game Players and Decision Variables

18.4.1.2 Solving the Grid Users' Game

18.4.1.3 Game Cost Functions

18.4.1.4 Task Execution Cost

18.4.1.5 Resource Utilization Cost

18.4.1.6 Security-Assurance Cost

18.4.2 Genetic-Based Resolution Methods for Game Models

18.4.2.1 Schedulers Implemented in Global Module

18.4.2.2 Local Schedulers in Players' Module

18.4.3 Empirical Setup

18.4.3.1 Performance Measures

18.4.4 Results

18.5 Conclusion

References

18.1 Introduction

The concept of the modern grid environment has grown far beyond the original Foster's grid idea of simple extension of conventional power networks. Today's grids are not just ...