Modeling and Analysis of Upload Utilization in BitTorrent System

Article Preview

Abstract:

Upload utilization is one of the most significant performance metrics in BitTorrent system, whereas there has been no accurate model which is suitable for it yet. For this reason, the delicate relation between some key parameters (maximum concurrent uploads, upload bandwidth, TFT round etc.) in BitTorrent is investigated, furthermore the piece diffusion model and the upload utilization model are proposed. Based on these models, we explore the effect of TFT round for the first time and find that the magnitude of TFT round has a significant influence on the transient stage as well as the end stage. Additionally, we show that when the seed’s upload bandwidth is fully utilized, to achieve optimal performance, each peer’s upload bandwidth should be proportional to its maximum concurrent uploads, and the file size should be large enough compared with the swarm size.

You have full access to the following eBook

Info:

Periodical:

Pages:

230-235

Citation:

Online since:

September 2011

Export:

Share:

Citation:

[1] B. Cohen, Incentives build robustness in BitTorrent, Proc. of the 1st Workshop on Economics of Peer-to-Peer Systems. Berkeley, (2003).

Google Scholar

[2] D. Qiu and R. Srikant, Modeling and Performance Analysis of BitTorrent-Like Peer-to-Peer Networks, Proc. of SIGCOMM'04, New York, 2004, p.367–378.

DOI: 10.1145/1030194.1015508

Google Scholar

[3] X. Yang and G. De Veciana, Performance of peer-to-peer networks: Service capacity and role of resource sharing policies, Performance Evaluation, vol. 63, 2006, pp.175-194.

DOI: 10.1016/j.peva.2005.01.005

Google Scholar

[4] D. Defigueiredo, B. Venkatachalam and S. F. Wu, Bounds on the performance of P2P networks using tit-for-tat strategies, Proc. of P2P'07, Galway, 2007, pp.11-18.

DOI: 10.1109/p2p.2007.32

Google Scholar

[5] R. Thommes and M. Coates, BitTorrent fairness: analysis and improvements, Proc. of 4th Workshop Internet, Telecom. and Signal, Noosa Heads, (2005).

Google Scholar

[6] P. Marciniak, N. Liogkas, A. Legout and E. Kohler, Small is not always beautiful, Proc. IPTPS'08, Berkeley, (2008).

Google Scholar

[7] X. Chen and S. A. Jarvis, Analysing bittorrent's seeding strategies, Proc. of 12th IEEE International Conference on Computational Science and Engineering, Vancouver, 2009, pp.140-149.

Google Scholar

[8] A. Bharambe, C. Herley, and V. Padmanabhan, Understanding and deconstructing BitTorrent performance, Microsoft Corp., Redmond, WA, Tech. Rep. MSR-TR-2005-03, (2005).

Google Scholar

[9] Information on http: /planete. inria. fr/software/BitSim.

Google Scholar