Font Size: a A A

Research On QoS-Based Anycast Routing Algorithms

Posted on:2007-10-07Degree:MasterType:Thesis
Country:ChinaCandidate:J J ZhouFull Text:PDF
GTID:2178360185977599Subject:Operational Research and Cybernetics
Abstract/Summary:PDF Full Text Request
With the rapid progress of the network and computer technologies, network service requirements exceed the capabilities of the network, which causes a great influence to application of Qulity of Service(QoS). In order to enhance the service ability and improve network flow distribution, we usually duplicate servers in the network. In recent years, people raised the communication model of "anycast" to research this kind of network service. Anycast has numerous potential applications, and is recognized as a useful service .More and more applications are related to anycast, so anycast is a standard service model defined by IPv6.In this dissertation we first generalize the development of the anycast communication service, and introduce the basic theory of QoS route. And also we propose two algorithms on based of current algorithm of QoS anycast route: First, the anycast route algorithm which is based on the basis of SSP (Shortest Shortest Path) with the constraint of the delay, bandwidth and its cost; Second, the anycast route algorithm based on genetic algorithm.In the first algorithm, we select the route according to the ratio of bottleneck bandwidth of each route to the product of the minimum time delay and the cost to make the network load statuesque and reduce the cost of network. In the second one, we consider the delay and the cost, and utilize a new fitness function to the compromise of the delay and the cost of service by adjusting parameters and selecting the route according to the request. The second algorithm has the properties of genetic algorithm such as good robustness, parallel search and population optimization.In order to demonstrate the algorithms' feasibility, we give out two simulations. The results of the experiments show: the delay is close to the SSP algorithm' and the cost in the first algorithm is lower than SSP algorithm' apparently; in the second algorithm, we propose route choosing operator to avoid being immersed into local solution and use node sequential coding to make the genetic operation easier and more effective. And also the proposed fitness...
Keywords/Search Tags:anycast, QoS(Quality of Service), routing algorithm, genetic algorithm
PDF Full Text Request
Related items