Dynamically Optimize Multicast Tree Structure
Ming Zeng, Hongliang Yu, Feng Yang, Weimin Zheng
Abstract
Ming Zeng, Hongliang Yu, Feng Yang, Weimin Zheng
Abstract
Live-streaming applications based on overlay multicast have become very popular recently. There have been many studies on optimizing the overlay construction and improving multicast performance. Most of these studies focused on building period of overlay network or multicast tree. It is not flexible in the heterogeneous and variable Internet environment. In this paper, we present DOMT, an overlay multicast protocol in which the multicast tree can be optimized dynamically. In DOMT, high capability peers get better position in the tree and more peers get near peers as their parents. Most of the previous protocols didn 't consider dynamic optimization because its bad impact to stability. In DOMT, we design some mechanisms to avoid the bad impact. The experimental results show that the optimization can decrease the height of multicast tree and shorten the average source-to-end delay. It also proves that the bad impact made by the optimization is acceptable
OpenAlex reports 3 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Live-streaming applications based on overlay multicast have become very popular recently. There have been many studies on optimizing the overlay construction and improving multicast performance. Most of these studies focused on building period of overlay network or multicast tree. It is not flexible in the heterogeneous and variable Internet environment. In this paper, we present DOMT, an overlay multicast protocol in which the multicast tree can be optimized dynamically. In DOMT, high capability peers get better position in the tree and more peers get near peers as their parents. Most of the previous protocols didn 't consider dynamic optimization because its bad impact to stability. In DOMT, we design some mechanisms to avoid the bad impact. The experimental results show that the optimization can decrease the height of multicast tree and shorten the average source-to-end delay. It also proves that the bad impact made by the optimization is acceptable
Key concepts: Multicast, Overlay multicast, Protocol Independent Multicast, Xcast, Computer science, Source-specific multicast, Pragmatic General Multicast, Computer network