Quality of Service in Next Generation High Speed Networks

Abstract

Named Data Networking (NDN) is a recently proposed new Future Internet architecture as an extension of Information Centric Networking defined in rfc-7476, rfc-7927. By naming data instead of locations, it changes the very basic network service abstraction from location-based delivery to name-based delivery. This fundamental change creates an abundance of new opportunities as well as many intellectual challenges in application development, network routing and forwarding, caching, security and privacy. The focus of this thesis is to design optimal forwarding and caching strategies. newlineThe research on forwarding focus on identifying working path for forwarding new or retransmitted interest. The forwarding strategies proposed either try to use single interface each time or perform a flooding on all available interfaces. Another important problem associated with forwarding is use of multiple paths for forwarding interest. In this thesis we propose maximum deviation based optimal forwarding strategy (MDOF) using Ordered Weighted Averaging Operator. Interest forwarding is modeled as Multi-attribute decision-making problem. Interest forwarding decision is then based on the ranking of interfaces according to the maximum deviation. newlineThe NDN routers caches all the content en-route to delivery. This makes user content available from any intermediate routers. This results in lowering load on content servers, reduced redundant transmission, no or little congestion on network path and decreased response time for fetching content for user requests. Caching all content is unpractical. En-route caching in all routers wastes the cache space. Early cache replacement may result in cache miss for same content requested by other users and router may need to run cache replacement operations often. Some content may never be requested by any users. Thus, an efficient and better caching strategy is needed for NDN. In this thesis, we have proposed three different caching strategies. newlinev newlineWe present low complexity, efficient cache placement strateg