An On-Demand QoS Routing Algorithm in Multiservice Scenarios
The trend that Internet applications are becoming more various puts forward higher request on Quality of Service (QoS) of coexisting multiservices in modern networks. An on-demand QoS path selection algorithm (marked as BW-cost algorithm) satisfying const
- PDF / 1,772,166 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 60 Downloads / 183 Views
An On-Demand QoS Routing Algorithm in Multiservice Scenarios Yanjing Li and Li Li
Abstract The trend that Internet applications are becoming more various puts forward higher request on Quality of Service (QoS) of coexisting multiservices in modern networks. An on-demand QoS path selection algorithm (marked as BWcost algorithm) satisfying constrained conditions of both bandwidth and cost is proposed innovatively in this paper, and its usage procedures in OSPF is claimed accordingly. The core idea of the algorithm is choosing the best route satisfying certain bandwidth requirement with smallest cost in order to achieve the goal of splitting streams and balancing load. Simulation results shows the BW-cost algorithm can successfully guarantee multiservices’ QoS demands as well as has much better performance in some targets such as throughput and network utility, comparing to traditional Dijkstra algorithm. It also exceeds en existing on-demand QoS algorithm. Keywords OSPF
Qos routing Dijkstra algorithm Multiservice scenario
15.1 Introduction A multiservice scenario has become the mainstream study scene in today’s networks as IP networks are demanded for more satisfying data services. Routing protocols which are using traditional algorithms to provide best-effort services can no longer meet the user’s QoS needs any more. For instance of the Dijkstra algorithm used in OSPF protocol, which is one of the most classic shortest path Y. Li L. Li (&) Xitucheng 10, Haidian District, Beijing, China e-mail: [email protected]
Z. Zhong (ed.), Proceedings of the International Conference on Information Engineering and Applications (IEA) 2012, Lecture Notes in Electrical Engineering 219, DOI: 10.1007/978-1-4471-4853-1_15, Ó Springer-Verlag London 2013
113
114
Y. Li and L. Li
first algorithms, it always leads to the overuse of calculated paths with minimum hop counts and consequently the congestion at this best path when some highquality-demanding data services trying to pass through. So some QoS mechanisms must be used to balance the load of the network and improve the transmission efficiency. QoS routing has been studied for a while. In [1] a cheapest path algorithm from one source to all destinations when links have two weights (cost and delay) is presented. Some other solutions use source routing along with shortest path routing to achieve the goal [2]. Anirudha presented a load sensitive routing (LSR) algorithm based on Dijkstra’s shortest path algorithm [2]. In [3], Karima et al. claimed that MPLS and traffic engineering provide indeed an adequate mean to establish constrained routes which satisfy application requirements like bandwidth. A novel two-phase load balanced shortest path routing (LB-SPR) is proposed in [4], where each phase uses the standard SPR protocol. In [5], authors described a path precomputation selection algorithm. A feature of it is that in each iteration, the intermediate results must be stored in the QoS routing table, which inevitably increases the expenses in terms of storage. Another al
Data Loading...
