RESEARCH PAPER International Journal of Recent Trends in Engineering, Vol 1, No. 1, May 2009 11 Bi-Criteria Latency Optimization of Intra- and Inter-Autonomous System Traffic Engineering Vidya. K 1 and Rhymend Uthariaraj. V. 2 1 Department of Computer Science and Engineering, Anna University Chennai, India Email: vidyapradeep21@yahoo.co.in 2 Ramanujan Computing Centre, Anna University Chennai, India Email: rhymend@annauniv.edu Abstract-Traffic Engineering (TE) is the process of controlling how traffic flows through a network in order to facilitate efficient and reliable network operations while simultaneously optimizing network resource utilization and traffic performance. TE improves the management of data traffic within a network and provides the better utilization of network resources. Many research works considers intra and inter Traffic Engineering separately. But in reality one influences the other. Hence the effective network performances of both inter and intra Autonomous Systems (AS) are not optimized properly. To achieve a better Joint Optimization of both Intra and Inter AS TE, we propose a joint Optimization technique by considering intra-AS features during inter – AS TE and vice versa. This work considers the important criterion say latency within an AS and between ASes and proposes a Bi-Criteria Latency optimization model. Hence an overall network performance can be improved by considering this joint-optimization technique in terms of Latency. Index terms- Inter-Domain, Intra-Domain, Traffic Engineering, Border Gateway Protocol, Latency, Optimization, Performance. I. I NTRODUCTION Traffic Engineering is one of the promising tools that provide QOS across the Internet. The need for Traffic Engineering includes Quality of Service (less latency, Higher through put, less packet loss, Higher Bandwidth), Interdependent tunable parameters, network growth, traffic variability and multicasting. These drive the need for better TE tools. Hence works on traffic Engineering focuses predominantly on Interior Gateway Protocols (IGPs) such as OSPF, IS-IS, and MPLS, which control the flow of traffic within an Autonomous system. But most traffic in a large backbone network traverses multiple domains, making interdomain traffic an important part of traffic Engineering.. Traffic engineering is performed by means of a set of techniques that can be used to better control the flow of packets inside an IP network. Here a small number of sources are responsible for a large fraction of the traffic. Across inter-domain boundaries, traffic engineering relies on a careful tuning of the route advertisements sent via the Border Gateway Protocol (BGP). This tuning can be used to control the flow of the incoming and of the outgoing traffic and identify their limitations. Several Internet Service Providers (ISPs) rely on traffic engineering to better control the flow of IP packets. Large ISPs often need to engineer the flow of packets inside their own domain to reduce congestion by better distributing the traffic on all their links. To remedy this situation, the techniques that tune the traditional IP routing protocols used inside the ISP network, is required. Besides optimizing the flow of packets inside their network, most ISPs also need to better control the flow of their inter-domain traffic, i.e. the IP packets that cross the boundaries between distinct ISPs. Internet Service Providers typically carry both intra –AS traffic that is routed only within their networks and inter-AS traffic that is routed not only within their networks but also across other ASes. The Border Gateway Protocol is the current de facto standard inter-domain routing protocol. Hence the only solution to engineer the flow of inter-domain traffic is to tune the configuration of the BGP routing protocol. However this tuning has its own limitations. Our Internet, which is an interconnection of multiple networks called domains or Autonomous Systems is composed of more than 25,000 ASes that collectively advertise more than 200,000 IPv4 prefixes, and these statistical numbers are still increasing. This interdomain routing system is one of the largest distributed systems today. Each domain should be able to select and distribute to its peers, the best path to reach each destination. The autonomous systems interact and coordinate the IP traffic delivery, exchanging routing information using this BGP protocol. This BGP is a complex protocol which enforces various economical relationships among domains. Previous research works on intra-AS traffic Engineering has assumed that ingress and egress points of inter-AS traffic do not change. Also prior inter-AS traffic does not consider the optimization of routes inside an AS. © 2009 ACADEMY PUBLISHER
