Computing Networks: From Cluster to Cloud Computing

ISBN-13: 9781848212862 / Angielski / Twarda / 2011 / 264 str.

Pascale Vicat-Blanc Primet; Pascale Vicat-Blanc Primet

Opis

-Computing Networks- explores the core of the new distributed computing infrastructures we are using today: the networking systems of clusters, grids and clouds. It helps network designers and distributed-application developers and users to better understand the technologies, specificities, constraints and benefits of these different infrastructures' communication systems.

Cloud Computing will give the possibility for millions of users to process data anytime, anywhere, while being eco-friendly. In order to deliver this emerging traffic in a timely, cost-efficient, energy-efficient, and reliable manner over long-distance networks, several issues such as quality of service, security, metrology, network-resource scheduling and virtualization are being investigated since 15 years. -Computing Networks- explores the core of clusters, grids and clouds networks, giving designers, application developers and users the keys to better construct and use these powerful infrastructures.

Spis treści

Introduction 13

Chapter 1. From Multiprocessor Computers to the Clouds 21

1.1. The explosion of demand for computing power 21

1.2. Computer clusters 24

1.3. Computing grids 26

1.4. Computing in a cloud 32

1.5. Conclusion 36

Chapter 2. Utilization of Network Computing Technologies 39

2.1. Anatomy of a distributed computing application 39

2.2. Programming models of distributed parallel applications 52

2.3. Coordination of distributed resources in a grid 57

2.4. Conclusion 60

Chapter 3. Specificities of Computing Networks 63

3.1. Typology of computing networks 63

3.2. Network transparency 68

3.3. Detailed analysis of characteristics expected from protocols 78

3.4. Conclusion 85

Chapter 4. The Challenge of Latency in Computing Clusters 87

4.1. Key principles of high–performance networks for clusters 88

4.2. Software support for high–performance networks 90

4.3. Description of the main high–performance networks 99

4.4. Convergence between fast and traditional networks 108

4.5. Conclusion 111

Chapter 5. The Challenge of Throughput and Distance 113

5.1. Obstacles to high rate 113

5.2. Operating principle and limits of TCP congestion control 115

5.3. Limits of TCP over long distances 120

5.4. Configuration of TCP for high speed 122

5.5. Alternative congestion–control approaches to that of standard TCP 126

5.6. Exploration of TCP variants for very high rate 133

5.7. Conclusion 136

Chapter 6. Measuring End–to–End Performances 139

6.1. Objectives of network measurement and forecast in a grid 139

6.2. Problem and methods 144

6.3. Grid network–performance measurement systems 155

6.4. Performance forecast 164

6.5. Conclusion 170

Chapter 7. Optical Technology and Grids 171

7.1. Optical networks and switching paradigms 172

7.2. Functional planes of transport networks 179

7.3. Unified control plane: GMPLS/automatic switched transport networks 184

Chapter 8. Bandwidth on Demand 189

8.1. Current service model: network neutrality 190

8.2. Peer model for bandwidth–delivery services 194

8.3. Overlay model for bandwidth–providing services 196

8.4. Bandwidth market 200

8.5. Conclusion 201

Chapter 9. Security of Computing Networks 203

9.1. Introductory example 203

9.2. Principles and methods 205

9.3. Communication security 212

9.4. Network virtualization and security 213

9.5. Conclusion 216

Chapter 10. Practical Guide for the Configuration of High–speed Networks 217

10.1. Hardware configuration 218

10.2. Importance of the tuning of TCP parameters 221

10.3. Short practical tuning guide 222

10.4. Use of multi–flow 226

10.5. Conclusion 228

Conclusion: From Grids to the Future Internet 229

Bibliography 235

Acronyms and Definitions 251

Index 263