Comprehensive overview of recent research advancements in scheduling approaches for cloud edge computing systems
Intelligent Scheduling of Tasks for Cloud-Edge-Device Computing Systems offers an in-depth collection of advanced task scheduling algorithms designed specifically for diverse cloud-edge-device computing systems. After an introductory overview, a series of intelligent scheduling approaches are presented, each specifically designed for a particular scenario within cloud-edge-device computing systems.
The book then summarizes the authors’ research findings in recent years, delving into topics including resource management, latency and real-time requirements, load balancing, priority constraints, algorithm design, and performance evaluation. The book enables readers to achieve efficient allocation of computing, storage, and network resources to optimize resource utilization. Real-world applications of scheduling technologies in smart cities and traffic management, industrial automation and smart factories, and healthcare monitoring systems are given in a separate chapter.
Additional topics include:
- Workload-aware scheduling of real-time independent tasks, covering how to schedule jobs in a single or multiple servers
- Mixed real-time task scheduling in automotive systems with vehicle networks, covering hybrid schedule design, offline task management, and online job assignment
- Scheduling with real-time constraint, covering task placement adjustment strategy, start time adjustment, and backwards schedule adjustment
- Energy-efficient scheduling without real-time constraint, covering energy consumption-optimal task placement plans as well as partition scheduling
Intelligent Scheduling of Tasks for Cloud-Edge-Device Computing Systems is an essential resource for researchers and practitioners in the field of IoT seeking to understand specific challenges and requirements associated with task scheduling in cloud-edge-device computing systems.
Table of Contents:
Foreword ix
About the Authors xi
Preface xv
Acknowledgments xvii
Acronyms xix
Glossary xxi
1 Introduction 1
1.1 Cloud-edge-device Computing Systems 1
1.2 Tasks 2
1.3 Task Scheduling 3
1.4 Outline of the Book 4
1.5 Summary 5
2 Scheduling Mixed Real-time Tasks in an Automotive System with Vehicular Network 7
2.1 Introduction 8
2.2 Related Work 10
2.3 Models and Problem Formulation 12
2.3.1 Software Model 12
2.3.2 Hardware Model 12
2.4 Hybrid Scheduler Design 13
2.5 Schedulability Test 15
2.5.1 Utilization-based Schedulability Test 15
2.5.2 Demand-supply Analysis 15
2.6 Offline Task Assignment 17
2.6.1 Problem Formulation 17
2.6.2 Hard Real-time Task Assignment 20
2.6.3 Soft Real-time Task Assignment 22
2.6.4 Complexity Analysis 24
2.7 Online Job Assignment 24
2.7.1 Online Schedulability Test 24
2.7.2 Job Assignment Strategy 27
2.7.3 Complexity Analysis 28
2.8 Performance Evaluation 28
2.8.1 Compared Approaches 29
2.8.2 Schedulability Test Results 30
2.8.3 Online Job Assignment Tests 33
2.9 Summary 35
3 Workload-aware Scheduling of Real-time Independent Tasks in Cloud 37
3.1 Introduction 37
3.2 Related Work 41
3.3 Related Models 43
3.3.1 Virtual CPU Model 43
3.3.2 Real-time Job Model 43
3.3.3 Power Model of VM 44
3.4 Problem Formulation 44
3.4.1 Input 45
3.4.2 Output 45
3.4.3 Constraints 45
3.4.4 Objective 46
3.5 Scheduling Jobs in a Single Server 47
3.5.1 Power Analysis 47
3.5.2 Problem Transformation 47
3.5.3 Dynamic Programming 48
3.6 Scheduling Jobs in Multiple Servers 51
3.6.1 Server Energy Efficiency 51
3.6.2 Job Placement in Multiple Servers 52
3.7 Online Workload-aware Scheduling 53
3.7.1 Job Frequency Profile 54
3.7.2 Energy-efficient Job Accommodation Scheme 54
3.8 Performance Evaluation 58
3.8.1 Simulation Setup 58
3.8.2 Compared Approaches 58
3.8.3 Results 59
3.9 Summary 64
4 Energy-minimized Scheduling of Real-time Dependent Tasks in Cloud 65
4.1 Introduction 65
4.2 Related Work 68
4.3 Problem Formulation 69
4.3.1 Input 69
4.3.2 Output 70
4.3.3 Objective 71
4.3.4 Constraints 71
4.4 Energy-efficient Scheduling Without Real-time Constraint 74
4.4.1 Energy Consumption-minimized Task Placement Plan 74
4.4.2 Partition Scheduling 75
4.5 Scheduling with Real-time Constraint 78
4.5.1 Task Placement Adjustment Strategy 79
4.5.2 Start Time Adjustment 82
4.5.3 Schedule Adjustment in a Backward Way 84
4.6 Performance Evaluation 85
4.6.1 Simulation Setup 86
4.6.2 Compared Approaches 87
4.6.3 Results 88
4.7 Summary 93
5 Workload-aware Scheduling of Real-time Dependent Tasks in Vehicular Edge Computing 95
5.1 Introduction 95
5.2 Related Work 97
5.3 Models and Problem Formulation 99
5.3.1 Vehicular Computing Model 99
5.3.2 Application Model 100
5.3.3 Power Model 101
5.3.4 Response Time Model 102
5.3.5 Problem Formulation 102
5.4 Decentralized Auction-bid Scheduling Scheme 103
5.4.1 Auction-bid Strategy 103
5.4.2 Task Prioritization 104
5.4.3 Task Assignment and Execution 104
5.4.4 Power Management 106
5.5 GS Scheme 107
5.5.1 Task Execution of Multiple Applications 107
5.5.2 Application Group and Allocation 108
5.6 Evaluation 110
5.6.1 Simulation Setup 110
5.6.2 Performance Results 111
5.7 Summary 117
6 Scheduling Multiple-criticality Dependent Tasks in Vehicular Edge Computing System 119
6.1 Introduction 119
6.2 Related Work 122
6.3 Problem Formulation 123
6.3.1 Input 123
6.3.2 Output 125
6.3.3 Constraints 125
6.4 Response Time Analysis 126
6.4.1 Task’s Response Time in a VM 127
6.4.2 Application’s Response Time 127
6.5 Scheduling at Level-1 Mode 128
6.5.1 Application Decomposition 128
6.5.2 State-transition Equation 129
6.5.3 Dynamic Programming 131
6.6 Mixed-criticality Scheduling 132
6.6.1 Mixed-criticality Schedulability Test 132
6.6.2 Online Management by Frequency Prediction 133
6.7 Performance Evaluation 134
6.7.1 Compared Approaches 134
6.7.2 Results 135
6.8 Summary 139
7 Real-world Applications of Scheduling Technologies 141
7.1 Introduction 141
7.2 Traffic Management 141
7.3 Smart Agriculture with Internet of Things 143
7.4 Healthcare Monitoring Systems 144
7.5 Summary 145
8 Summary and Future Research 147
8.1 Summary 147
8.2 Future Research 147
References 149
Index 165
About the Author :
Biao Hu is an Associate Professor with the College of Engineering at China Agricultural University, Beijing, China.
Mingguo Zhao is a Professor with the Department of Automation at Tsinghua University, Beijing, China.
Zhengcai Cao is a Professor with the School of Mechatronics Engineering at Harbin Institute of Technology, Harbin, China.
Mengchu Zhou is a Professor with the Helen and John C. Hartmann Department of Electrical and Computer Engineering at the New Jersey Institute of Technology, Newark, NJ, USA.
