Towards Minimizing Average Finish Time of P2P File Delivery Under Peer Leaving

Optimized Water Filling scheduling (OWF) has been studied previously \cite{p2p:Ezovski2009} for no peer leaving Peer-to-Peer (P2P) network, where the problem of efficiently delivering a common set of data from a single base station to multiple users that have heterogeneous uploading capacity was examined. In the absence of peer leaving, it has been demonstrated that OWF can achieve significant performance improvement by exploiting the optimal tradeoff between augment of last finish time and reduction of average finish time. However, peer leaving is a typical behavior of P2P file sharing network, and minimizing average finish time combined with the effects of peer leaving necessitates a more complex scheduling policy. In this work, we extend their studies to P2P networks with users are subject to leave when they complete the data gathering. A simple scheduling method, named reversed water filling method (RWF) is proposed to adapt problem of peer leaving. We also examine the finish time and total network access time of peers under different file distribution strategies. The results show that RWF method yields significant shorter average finish time and total network access time than its predecessors and all other known file delivery methods under peer leaving condition. The results provide fundamental insights of scheduling for improving performance of P2P systems.

[1]  Jeffrey G. Andrews,et al.  Comparison of space-time water-filling and spatial water-filling for MIMO fading channels , 2004, IEEE Global Telecommunications Conference, 2004. GLOBECOM '04..

[2]  Ya Zhou,et al.  A Request Handling Mechanism with the Shortest Processing-Time First in Web Servers , 2008, 2008 International Conference on Computer Science and Software Engineering.

[3]  Han Hoogeveen,et al.  Minimizing the number of late jobs in a stochastic setting using a chance constraint , 2008, J. Sched..

[4]  Bin Fan,et al.  The design trade-offs of BitTorrent-like file sharing protocols , 2009, TNET.

[5]  Eytan Modiano Random algorithms for scheduling multicast traffic in WDM broadcast-and-select networks , 1999, TNET.

[6]  E.L. Lawler,et al.  Optimization and Approximation in Deterministic Sequencing and Scheduling: a Survey , 1977 .

[7]  K. Shibata,et al.  Optimum scheduling in pipeline processing , 1992, [Proceedings] Singapore ICCS/ISITA `92.

[8]  Gustavo de Veciana,et al.  Service capacity of peer to peer networks , 2004, IEEE INFOCOM 2004.

[9]  Gideon Weiss,et al.  Analysis of peer-to-peer file dissemination amongst users of different upload capacities , 2006, PERV.

[10]  Jeffrey D. Ullman,et al.  NP-Complete Scheduling Problems , 1975, J. Comput. Syst. Sci..

[11]  B. Cohen,et al.  Incentives Build Robustness in Bit-Torrent , 2003 .

[12]  Li Wang,et al.  A scheduling algorithm for flexible flow shop problem , 2002, Proceedings of the 4th World Congress on Intelligent Control and Automation (Cat. No.02EX527).

[13]  K. V. Shakhbazyan,et al.  A survey of scheduling methods for multiprocessor systems , 1981 .

[14]  Rayadurgam Srikant,et al.  Modeling and performance analysis of BitTorrent-like peer-to-peer networks , 2004, SIGCOMM 2004.

[15]  Wooseung Jang,et al.  Scheduling unrelated parallel machines to minimize total weighted tardiness , 2006, SOLI 2006.

[16]  Sanjay Ranka,et al.  Using game theory for scheduling tasks on multi-core processors for simultaneous optimization of performance and energy , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[17]  Daniel Stutzbach,et al.  Understanding churn in peer-to-peer networks , 2006, IMC '06.

[18]  Peter Brucker,et al.  Scheduling Algorithms , 1995 .

[19]  Lachlan L. H. Andrew,et al.  Minimizing Average Finish Time in P2P Networks , 2009, IEEE INFOCOM 2009.

[20]  Kam-Wing Ng,et al.  Modeling, Analysis and Improvement for BitTorrent-Like File Sharing Networks , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[21]  Peter A. Dinda,et al.  Improving peer-to-peer performance through server-side scheduling , 2008, TOCS.

[22]  Ayalvadi J. Ganesh,et al.  Resource allocation between persistent and transient flows , 2005, IEEE/ACM Transactions on Networking.

[23]  S.G. Ponnambalam,et al.  Hoprield neural network approach for single machine scheduling problem , 2004, IEEE Conference on Cybernetics and Intelligent Systems, 2004..

[24]  Bin Fan,et al.  The Delicate Tradeoffs in BitTorrent-like File Sharing Protocol Design , 2006, Proceedings of the 2006 IEEE International Conference on Network Protocols.

[25]  Yun Ling,et al.  A Dependable Cluster Based Topology in P2P Networks , 2010, J. Commun..

[26]  Rakesh Kumar,et al.  Peer-Assisted File Distribution: The Minimum Distribution Time , 2006, 2006 1st IEEE Workshop on Hot Topics in Web Systems and Technologies.

[27]  Chuan Wu,et al.  Multi-Channel Live P2P Streaming: Refocusing on Servers , 2008, IEEE INFOCOM 2008 - The 27th Conference on Computer Communications.

[28]  Ronald L. Graham,et al.  Bounds on Multiprocessing Timing Anomalies , 1969, SIAM Journal of Applied Mathematics.

[29]  Tracey Ho,et al.  Optimal Strategies for Efficient Peer-to-Peer File Sharing , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[30]  Arun Venkataramani,et al.  Do Incentives Build Robustness in BitTorrent? (Awarded Best Student Paper) , 2007, NSDI.