Skip to main content
eScholarship
Open Access Publications from the University of California

UC San Diego

UC San Diego Electronic Theses and Dissertations bannerUC San Diego

Destination-based Routing and Circuit Allocation for Future Traffic Growth

Abstract

Internet traffic continues to grow relentlessly, driven largely by increasingly high- \ resolution video streaming, the increasing adoption of cloud computing, the emergence of 5G networks, and the ever-growing reach of social media and social networks. Existing networks use packet switching to route packets on a hop-by-hop basis from the source to the destination. However, they suffer from two shortcomings. First, in existing networks, packets are routed along a fixed shortest path using the Open Shortest Path First (OSPF) protocol or obliviously load-balanced across equal-cost paths using the Equal-Cost Multi-Path (ECMP) protocol. These routing protocols do not fully utilize the network capacity because they do not adapt to network congestions in their routing decisions. Second, although studies have shown that the majority of packets processed by Internet routers are pass-through traffic, packets nonetheless have to be queued and routed at every hop in existing networks, which unnecessarily adds substantial delays and processing costs.

In this thesis, we present two new approaches to overcome these shortcomings. First, we propose new backpressure-based routing algorithms which use only shortest-path routes when they are sufficient to accommodate the given traffic load, but will incrementally expand routing choices as needed to accommodate increasing traffic loads. This avoids the poor delay performance inherent in backpressure-based routing algorithms where packets may take long detours under light or moderate loads, and still retains the notable advantage, the network-wide optimal throughput, because packets are adaptively routed along less congested paths.

Second, we propose a unified packet and circuit switched network in which the underlying optical transport is used to circuit-switch pass-through traffic by means of pre-established circuits. This avoids unnecessary packet queuing delays and processing costs at each hop. We propose a novel convex optimization framework based on a new destination-based multicommodity flow formulation for the allocation of circuits in such unified networks.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View