Header menu link for other important links
X
Stabilizing the performance and fairness in dynamic p2p system using Do or Die incentive mechanism
, R. Asha, P. Anandhakumar
Published in Institute of Electrical and Electronics Engineers Inc.
2015
Pages: 229 - 236
Abstract
Live streaming applications for Video conferencing, E-learning and Telemedicine have become popular over the Internet. Peer-to-Peer network best suites the streaming architecture because p2p distributed network comprises of nodes which communicates with its neighboring nodes to make chunks available to other peer in the network environment. p2p paradigm serves as the effective solution in delivering videos with cost effective infrastructure. 'Free Rider' one of the phenomenon which is identified as one of the challenge in the p2p system. These Free Riders are nothing but the node which simply resides in the p2p network, downloads the incoming streams but refuse to contribute or upload the file to other nodes. This serious threat degrades the performance of the overall system. The stream that it keeps downloading will rapidly increase leaving the system heavily loaded. On the other hand, the contributing nodes will be flooded with queries requesting for the streams. This leads to unfairness in the p2p system. This paper proposes an incentive mechanism 'Do or Die' which boosts up the contributing nodes, as a sign of encouragement for their contribution, at the same time simultaneously motivates and threatens the Free Riders to contribute, else leave the network. The proposed incentive mechanism works on increment/Decrement point based scheme. Here the individual nodes are given increment points for uploading the file and decrement point for downloading the file. Hence the contributing nodes acquire high points on uploading and the Free Riders gets their points decreased as they only download the file. Hence the Free Riders can be identified and eliminated based on their decrement point. As the no of Free Riders decreases the overall system performance increases. Simulation results prove that the percentage of Free Riders reduced from 60% to 51% in the system and thereby increasing the performance and fairness of the system. © 2014 IEEE.
About the journal
JournalData powered by Typeset6th International Conference on Advanced Computing, ICoAC 2014
PublisherData powered by TypesetInstitute of Electrical and Electronics Engineers Inc.