Huagang Shao

An efficient qos framework with distributed adaptive resource management in IPv6 networks

In this paper, we proposed a new QoS framework with Distributed Adaptive Resource Management(DARM). DARM provides end-to-end QoS guarantees to individual flows with minimal overhead, while keeping the scalability characteristic of DiffServ. In DARM, per-flow admission control and resource reservation, in conjunction with a novel IPv6 flow label mechanism, can be processed instantaneously in a fully distributed and independent manner at edge of network without hop-by-hop signaling. In addition, DARM is capable of reconfiguring network resource adaptively according to dynamically changing of traffic load. Through extensive simulations, the results clearly exhibit that DARM has a better overall performance comparing to the IntServ and DiffServ.

A Multiobjective Optimization Algorithm for LSP Setup in Diffserv and MPLS Networks

In this paper, a multiobjective optimization QoS-aware path selection algorithm is proposed for LSPs setup in DiffServ-MPLS networks. We simultaneously optimize multiple QoS objectives by a genetic algorithm in conjunction with concept of Pareto dominance. Simulation demonstrates that the proposed algorithm is capable of discovering a set of QoS-aware near optimal paths within in a few iterations. In addition, simulation results also very promising indicates scalability of the algorithm with increasing number of network nodes.

DIRA: distributed insertion and relocation routing algorithm for overlay multicast in diffserv domain

Traditional multicast routing algorithms require routers to be specially designed such that they can forward the same copy of packet onto different output links simultaneously. This idea conflicts severely with the core-stateless principle in most QoS network architectures, such as Diffserv. Meanwhile, overlay multicast emerges as an effective alternative to the traditional approaches. In this paper, we combine the overlay multicast technique with the Diffserv architecture seamlessly and design an overlay multicast routing algorithm, which generates a multicast tree at a pretty low cost incrementally. Owing to its distributed nature, the computation of the tree can be executed efficiently. Through a large amount of simulations, we have shown that our algorithm is quite competitive in terms of network resource utilization and other metrics.

An end-to-end qos provisioning architecture in IPv6 networks

Differentiated Service (DiffServ) is a scalable solution to provide class-based differentiated Quality of Services (QoS). However, the basic DiffServ model lacks mechanisms to reserve resource for the class-based traffic and perform per-flow admission control. In this paper, we propose an end-to-end QoS provisioning architecture in IPv6 networks supported by a QoS-aware path selection algorithm. We design a novel flow label mechanism to achieve the effect of per-flow reservation paradigm and keep the scalability of DiffServ. In addition, the proposed QoS-aware path selection algorithm can optimize multiple QoS objectives simultaneously by exploiting genetic algorithm technique in conjunction with concept of Pareto dominance. The simulation results demonstrate the efficiency and scalability of the proposed architecture and algorithm.

A novel approach for topology-aware overlay multicasting

Most existing overlay multicast approaches avoid to consider any network layer support no matter whether it is available or not. This design principle greatly increases the complexity of the routing algorithms and makes the overlay topologies incompatible with the underlying network. To address these issues, we propose TOMIMN as a novel overlay multicast approach, which exploits the cooperation between end-hosts and IP multicast routers to construct a topology-aware overlay tree. Through a little modification to PIM-SM, a multicast router is able to receive registration from nearby group members and redirect passing-by join requests to them. Due to the multicast routers support, TOMIMN organizes its group members into an overlay multicast tree efficiently, which matches the physical network topology well.

An Efficient Routing Algorithm for Overlay Multicast in Diffserv Domains

We propose a novel approach to enable QoS-guaranteed overlay multicast among Diffserv domains. Our approach exploits the edge routers as the overlay nodes while complying with the core-stateless principle of the Diffserv architecture. The contribution of our work is two-fold. First, we devise a distributed insertion and relocation algorithm (DIRA) to organize the edge routers of the same domain into an efficient overlay multicast tree. The algorithm keeps a candidate parent for each node so that as the incremental tree building process is going on, the tree topology can be adjusted timely for a small cost according to its dynamic membership. Due to its distributed nature, the algorithm imposes no huge computation burden on any single node. Second, our approach combine the overlay multicast technique with the QoS architecture seamlessly, which provides a promising platform for various applications. Through a large amount of simulations, we prove that DIRA is more competitive than other IP multicast and overlay multicast schemes in terms of tree cost and other metrics.