A Modularized QoS Multicasting Approach on Common Homogeneous Trees for Heterogeneous Members in DiffServ
Suogang Li, Jianping Wu, Ke Xu, Ying Liu
Abstract
Suogang Li, Jianping Wu, Ke Xu, Ying Liu
Abstract
Traditional IP multicast suffers from forwarding state scalability problems as the number of concurrent active multicast groups increases. The problem is exacerbated when provisioning QoS since additional information of resource requirement from members must be kept at routers. In this paper, we firstly consider the comparability between various QoS levels and propose a modularized QoS multicasting approach in the DiffServ model. To achieve the multicast state scalability, the trees, called common trees, are decoupled from groups. The groups including the members requiring heterogeneous QoS are divided into subgroups and delivered through the common homogeneous QoS trees. We put forward a tree-choosing algorithm to find out appropriate trees for a group. Extensive simulations demonstrate that the approach is able to improve multicast forwarding state scalability as well as supporting different QoS requirement. The modularized approach can be considered a building block for QoS multicasting in the DiffServ architecture for its ease of deployment and compatibility with DiffServ
OpenAlex reports 2 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.
Traditional IP multicast suffers from forwarding state scalability problems as the number of concurrent active multicast groups increases. The problem is exacerbated when provisioning QoS since additional information of resource requirement from members must be kept at routers. In this paper, we firstly consider the comparability between various QoS levels and propose a modularized QoS multicasting approach in the DiffServ model. To achieve the multicast state scalability, the trees, called common trees, are decoupled from groups. The groups including the members requiring heterogeneous QoS are divided into subgroups and delivered through the common homogeneous QoS trees. We put forward a tree-choosing algorithm to find out appropriate trees for a group. Extensive simulations demonstrate that the approach is able to improve multicast forwarding state scalability as well as supporting different QoS requirement. The modularized approach can be considered a building block for QoS multicasting in the DiffServ architecture for its ease of deployment and compatibility with DiffServ
Key concepts: Multicast, Quality of service, Computer science, Computer network, Scalability, Distributed computing, Homogeneous, Tree (set theory)