Server and route selection method for QoS-based anycast protocol

Abstract

The anycast communication protocol is to select the best server and the best route for a particular client from a group of replicated servers with the same contents. As users have come to demand contents with high QoS, the necessity to support the anycast communications in QoS networks has grown. In this paper, we propose a server and route selection method with application-level QoS-based anycast protocol. The protocol we propose has the following advantages. 1) The latest server and route information can be constantly acquired through distributed resource management by E-BB. 2) The server and route selection algorithm enable us to consider both the server load and network load simultaneously. 3) High reliability can be guaranteed by decentralized control.

An integrated end-to-end QoS anycast routing on DiffServ networks

“…Regarding the work by Nakaniwa et al [9], the first two problems are the same as those found in the work by Hao et al [8] and mentioned above. In addition, the protocol broadcasts a search message to all candidate servers in order to find the best server.…”

“…Since Lin et al [10] balanced only the server loading and network traffic, Figs. 9-11 compare and evaluate four different approaches, namely those proposed by Hao et al [8], Nakaniwa et al [9], Lin et al [10] and the proposed protocol. The remaining figures compare and evaluate only three different approaches, namely those proposed by Hao et al [8], Nakaniwa et al [9] and the proposed protocol.…”

“…9-11 compare and evaluate four different approaches, namely those proposed by Hao et al [8], Nakaniwa et al [9], Lin et al [10] and the proposed protocol. The remaining figures compare and evaluate only three different approaches, namely those proposed by Hao et al [8], Nakaniwa et al [9] and the proposed protocol. In addition, five performances, namely request blocking rate, average server's response time, average number of active client sessions, average time to find a feasible path and server, and average time to compute the RPM and MRCM, are considered in this paper.…”

“…The work by Nakaniwa et al [9] proposes a new application-level QoS-based anycast protocol, which considers both the server load and the network load simultaneously. The protocol improves the system reliability by introducing distributed resource management by the bandwidth broker (BB) in each domain.…”

“…There is a small body of knowledge on QoS-based anycast routing [8][9][10]. The article by Hao, Zegura and Ammar [8] proposes a remarkable end-to-end QoS-based anycasting architecture which consists of four major issues, namely client demand prediction, signaling protocol between resolver, server agent (SA) and BB, server selection/sorting algorithms, and resource reservation granularity.…”