For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
A New Content-Oriented Traffic Engineering for Content Distribution: CAR (Content Aware Routing)
Shigeyuki YAMASHITA Daiki IMACHI Miki YAMAMOTO Takashi MIYAMURA Shohei KAMAMURA Koji SASAYAMA
IEICE TRANSACTIONS on Communications
Publication Date: 2015/04/01
Online ISSN: 1745-1345
Type of Manuscript: PAPER
Category: Network System
traffic engineering, traffic matrix, content-oriented routing,
Full Text: PDF(2.6MB)
>>Buy this Article
Large-scale content transfer, especially video transfer, is now a dominant traffic component in the Internet. Originally, content transfer had a content-oriented feature, i.e., “Users do not care where content is retrieved. Users only take care of what content they obtain.” Conventional traffic engineering (TE) aims to obtain optimal routes for traffic between ingress and egress router pairs, i.e., TE has focused on a location-oriented approach that takes care of where to connect. With increased demand for content-oriented features for content traffic, TE needs to focus on content-oriented routing design. In this study, we therefore propose a novel approach to content-oriented TE, called content aware routing (CAR). In CAR, routes are designed for content and egress router pairs, i.e., content traffic toward a receiver-side router. Content demand can be flexibly distributed to multiple servers (i.e., repositories) providing the same content, meaning that content can be obtained from anywhere. CAR solves the optimization problem of minimizing maximum link utilization. If there are multiple optimal solutions, CAR selects a solution in which resource usage is minimized. Using numerical examples formulated by the linear programming problem, we evaluated CAR by comparing it with combinations of conventional content delivery networks and TE, i.e., location-oriented designs. Our numerical results showed that CAR improved maximum link utilization by up to 15%, with only a 5% increase of network resource usage.