A hybrid computing approach to improve convergence time for scalable network

  • PDF / 1,126,215 Bytes
  • 11 Pages / 595.276 x 790.866 pts Page_size
  • 15 Downloads / 204 Views

DOWNLOAD

REPORT


Sådhanå (2020) 45:225 https://doi.org/10.1007/s12046-020-01454-7

Sadhana(0123456789().,-volV)FT3](012345 6789().,-volV)

A hybrid computing approach to improve convergence time for scalable network R N DEVIKAR1,*, D V PATIL2 and V CHANDRA PRAKASH1 1

Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur 522502, India 2 Department of Computer Engineering, Gokhale Education Society’s R. H. Sapat College of Engineering, Management Studies and Research, Nasik 422005, India e-mail: [email protected]; [email protected]; [email protected] MS received 26 September 2019; revised 12 June 2020; accepted 15 July 2020 Abstract. Border Gateway Protocol (BGP) is a widely used routing protocol in the new era for the intercommunication between the multiple autonomous systems and it has been largely on the internet in all categories of the scalable network. In the event of failure, the BGP as an inter-domain routing protocol shows slow convergence, which results in high considerable delay in several internet/web applications. The minimum route advertisement interval (MRAI) timers are mostly used by network operators to reduce the issues occurring at the time of increasing convergence time. Many researchers have been working on variation in MRAI timer and effect of it on scalability and network convergence. The increasing size of a network leads to an increase in the value of MRAI timers. Hence, keeping the value of MRAI timers optimum results in reducing the issue of slow convergence for the scalable network. The proposed system (FAPSO) reduces the problem of convergence time by incorporating fuzzy logic into Particle Swarm Optimization (PSO) algorithm for the scalable network. In comparison with the static value of MRAI timer i.e., 30 s, FAPSO is a suitable algorithm that gives the optimal value of convergence time for the scalable network. Keywords.

Convergence time; MRAI; FAPSO; inter-domain routing.

1. Introduction BGP (Border Gateway Protocol) is the backbone routing protocol of the internet used in inter-domain routing. In the internet, BGP is used to establish and maintain the connections between the autonomous systems. Many researchers have widely studied the BGP convergence property as BGP has unreasonable behaviour with respect to convergence [1–3] and [4]. BGP has different behaviour and working compared with other routing protocols as by using various routing policies it manages the routing paths effectively and more efficiently. The network topology is used to create routing information. This information is most essential to improve the performance of a network, routing policies and topology design. Fast convergence of a network can help to improve the fast rerouting, minimize end to end delay, and reduce packet losses. Network faults such as node/link failure or route flapping have a latency ranging from several seconds to several minutes. Such condition leads to loss of packets which in turn affect the network performance. Convergence time o