Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

SNAP: SNowbAll multi-tree Pushing for Peer-to-Peer Media Streaming (1009.2282v1)

Published 13 Sep 2010 in cs.NI

Abstract: Given the respective advantages of the two complimentary techniques for peer-to-peer media streaming (namely tree-based push and mesh-based pull), there is a strong trend of combining them into a hybrid streaming system. Backed by recently proposed mechanisms to identify stable peers, such a hybrid system usually consists of backbone trees formed by the stable peers and other overlay structures in the second tier to accommodate the remaining peers. In this paper, we embrace the hybrid push-pull structure for peer-to-peer media streaming. Our protocol is dominated by a multi-tree push mechanism to minimize the delay in the backbone and is complemented by other overlay structures to cope with peer dynamics. What mainly distinguishes our multi-tree pushing from the conventional ones is an unbalanced tree design guided by the so called snow-ball streaming, which has a provable minimum delay and can be smoothly "melded" with virtually any other existing overlay structures lying in the second tier. We design algorithms to construct and maintain our SNowbAll multi-tree Pushing (SNAP) overlay, and we also illustrate how to smoothly weld the SNAP backbone with the second tier. Finally, we perform simulations in ns-2; the results indicate that our approach outperforms a recently proposed hybrid streaming system.

Summary

We haven't generated a summary for this paper yet.