Editorial
Special section: Management and optimisation of P2P and Grid systems with network economics

https://doi.org/10.1016/j.future.2010.04.010Get rights and content

Introduction

Peer-to-Peer (P2P) systems and Grids started as two different distributed computing philosophies: Grid computing represented a more enterprise-wide or multi-institutional orchestration of relatively homogeneous and powerful distributed computing resources to automate and optimise the execution of time-consuming tasks, whereas P2P networks had an emphasis on the discovery and sharing of resources on the edges of a heterogeneous network (i.e. user desktops, etc.), especially for file sharing. Nowadays these two domains have a substantial overlap as P2P systems have been borrowing more traditional Grid techniques for task scheduling and execution and the Grid systems have been deploying P2P-oriented techniques for more scalable and versatile resource discovery [1]. In fact, from a business-oriented perspective, both technologies have to answer the same non-trivial dual question: what incentives are given to resource providers to join a particular Grid or P2P network and how can the system as a whole provide to its users a fair and scalable access to these resources? The challenge in addressing these questions is further compounded by the fact that any potential solution is bound to have an impact on other important system-wide parameters such as performance, data traffic and fault tolerance, and therefore it must be coupled with an overall system management and optimisation strategy.

Network economics have emerged as a candidate methodology that can answer both of the questions above [2]. The term generally refers to a class of principles and techniques reused from the field of economics and applied into a distributed system in order to control access to network resources. The mechanisms used typically range from imposing simple incentives for bringing resources into a network (rather than just using existing ones) all the way to establishing and enforcing a full economy incorporating revenue-making schemes and charging policies. Between these two extremes, there are a number of variants that typically utilise tokens or rewards in conjunction with corresponding penalty schemes. Regardless of their specific operation, these techniques need to be combined with an appropriate security layer that guards against any inappropriate or fraudulent generation, distribution and use of rewards and incentives [3].

As opposed to more traditional algorithmic approaches to P2P/Grid management and optimisation, network economics driven mechanisms are appealing for three good reasons. Firstly, these schemes are typically based on a few simple fundamental principles that can make the system operation understandable to both resource providers and consumers. Secondly, they help identify ways in which it can become profitable for big content providers and ISPs to join the system. Finally, the right combination of incentives/rewards with charging policies can have a direct effect on resource usage patterns and therefore can be used as an effective instrument to manage the whole system and keep it self-sustained without any central intervention.

This special section aims to address the need for more research into the applicability and effectiveness of network economics as a management and optimisation methodology for P2P and Grid systems. It includes nine high-quality papers selected after careful peer review from more than 35 submissions.

Section snippets

Special section contents

One issue that consistently attracts more interest in P2P network management is that of restricting free riding behaviour. Indeed free riding is shown to significantly impact the performance and overall usefulness of a typical network. In the paper Resisting Free Riding Behaviour in Bit Torrent, Wang et al. present a mechanism that guarantees fairness among peers by using a novel encrypted block trading protocol. Its key concept is that of peers “prepaying” for encrypted data blocks and

Acknowledgements

In closing, we would like to thank all the authors who submitted their work to this special section as well as the reviewers who, through their expert and insightful comments, helped improve the quality of the submitted material significantly. We would also like to thank Elsevier’s editorial team for their continued guidance and professional support throughout all the phases of this publication. We hope you will find the material in this special section of FGCS interesting and useful.

Nick Antonopoulos is currently the Head of School of Computing and Assistant Dean (Research) of the Faculty of Business, Computing and Law at the University of Derby. Prior to joining the University of Derby in March 2009, he was a Senior Lecturer (US Associate Professor) and Director of the M.Sc. Degrees at the Department of Computing, University of Surrey, UK. He holds a B.Sc. in Physics (1st class) from the University of Athens (1993), an M.Sc. in Information Technology from Aston University

First page preview

First page preview
Click to open first page preview

References (3)

There are more references available in the full text version of this article.

Cited by (0)

Nick Antonopoulos is currently the Head of School of Computing and Assistant Dean (Research) of the Faculty of Business, Computing and Law at the University of Derby. Prior to joining the University of Derby in March 2009, he was a Senior Lecturer (US Associate Professor) and Director of the M.Sc. Degrees at the Department of Computing, University of Surrey, UK. He holds a B.Sc. in Physics (1st class) from the University of Athens (1993), an M.Sc. in Information Technology from Aston University (1994), and a Ph.D. in Computer Science from the University of Surrey (2000). Prior to joining academia, he worked as a networks consultant and was the co-founder and director of a company developing Web-based management information systems. He has over 11 years of academic experience during which he has designed and has been managing advanced Masters Programmes in Computer Science at the University of Surrey. He has published over 80 articles in fully refereed journals and international conferences. He has received a number of best paper awards in conferences and graduated six Ph.D. students. He was the organiser and chair of the 1st international workshop on computational P2P networks (ComP2P 2008). He is currently the chair of the 1st international workshop on service-oriented P2P networks (ServP2P, CCGrid 2009) and the general chair of the international conference on advances in P2P systems (AP2PS 2009). He is on the editorial board of the Springer journal of Peer-to-Peer Networking and Applications (effective from 2009) and on the advisory editorial board of the IGI Global Handbook of Research on Telecommunications Planning and Management for Business. He is a Fellow of the UK Higher Education Academy and a full member of the British Computer Society. His research interests include emerging technologies such as large-scale distributed systems and peer-to-peer networks, software agent architectures and security. Contact him at the School of Computing, University of Derby, Derby, DE22 1GB, United Kingdom, [email protected].

Omer F. Rana is a Professor of Performance Engineering in the School of Computer Science at Cardiff University, and the Deputy Director of the Welsh eScience Centre. He holds a Ph.D. in “Neural Computing and Parallel Architectures” from Imperial College, London, and works in the areas of high-performance distributed computing, multi-agent systems and Data Mining. He has published over 150 papers in peer-reviewed, international journals, workshops and conferences. He participates on the Editorial boards of Concurrency and Computation: Practice and Experience (Wiley), Scientific Programming (IOS Press), the International Journal of Pervasive Computing (Troubador Press) and the Journal of Computational Science (Elsevier). He is an Associate Editor of the ACM Transactions on Autonomous and Adaptive Systems. He is on the editorial board of the new book series on Autonomic Systems (Birkhauser, Switzerland), and on the steering committee of the IEEE/ACM CCGrid conference series. His research has focused on the use of intelligent techniques for resource and performance management within distributed systems, in particular support for machine learning and adaptation mechanisms for supporting adaptive resource allocation. Prior to joining Cardiff University, he worked as a software developer with Marshall BioTechnology Limited (based in London), focusing on dynamic data analysis from scientific instruments. He worked on projects with Marshall Aerospace (Cambridge), Fisons Applied Sensor Technology (Cambridge), Amersham International (North London) and Hybaid (California). He can be reached at the School of Computer Science, Cardiff University, 5 The Parade, Cardiff CF24 3AA, Wales, UK, [email protected].

View full text