Equal-cost Multi-path Routing Research Articles

Enhanced MENTOR Algorithm with ECMP Routing Enable

MENTOR (Mesh Network Topological Optimization and Routing) is an efficient heuristic design algorithm network that assigns traffic flow while selecting links to be installed. MENTOR is suitable to design networks categorized as virtual circuit packet switching such as Frame Relay, ATM and MPLS networks. However, MENTOR cannot be directly applied to datagram packet switching networks such as purely IP router network. This is because the traffic flows determined by original MENTOR are not always the shortest path routing based on IP link weight assignment. To solve this problem, Cahn proposed a modified version of MENTOR called MENTOR-II in which appropriate link weights are assigned to the installed links such that the traffic always route on the path with minimum distance. However, it should be noted that the traffic flows assigned by MENTOR-II are limited to a single path routing. This study proposed a new modified version of MENTOR called Enhance-MENTOR, or EMENTOR, which is able to assign traffic with the Equal Cost Multi-Path (ECMP) routing. Three options of E-MENTOR are considered, i.e. E-MENTOR with single path routing only, E-MENTOR with ECMP routing only and E-MENTOR that consider ECMP routing if benefit. The three options of E-MENTOR are evaluated in term of installation cost, routing cost and reliability for 729 network design conditions. The design results are compared with those of original MENTOR and MENTOR-II.

On the efficacy of fine-grained traffic splitting protocolsin data center networks

Multi-rooted tree topologies are commonly used to construct high-bandwidth data center network fabrics. In these networks, switches typically rely on equal-cost multipath (ECMP) routing techniques to split traffic across multiple paths, such that packets within a flow traverse the same end-to-end path. Unfortunately, since ECMP splits traffic based on flow-granularity, it can cause load imbalance across paths resulting in poor utilization of network resources. More fine-grained traffic splitting techniques are typically not preferred because they can cause packet reordering that can, according to conventional wisdom, lead to severe TCP throughput degradation. In this work, we revisit this fact in the context of regular data center topologies such as fat-tree architectures. We argue that packet-level traffic splitting, where packets of a flow are sprayed through all available paths, would lead to a better load-balanced network, which in turn leads to significantly more balanced queues and much higher throughput compared to ECMP.

New strategy for optimizing wavelength converter placement

This paper proposes a new strategic alternate-path routing to be combined with the particle swarm optimization (PSO) algorithm to better solve the wavelength converters placement problem. The strategic search heuristic is designed to provide network connectivity topologies for the converters to be placed more effectively. The new strategy is applied to the 14-node NSFNET to examine its efficiency in reducing the blocking probability in sparse wavelength conversion network. Computed results show that, when applied to the identical optimization framework, our search method outperforms both the equal-cost multipath routing and traffic-engineering-aware shortest-path routing.

Enhancing wavelength converter placement optimization with traffic-engineering-aware shortest-path routing

This paper proposes the combination of multiple shortest-paths routing and particle swarm optimization algorithm to solve the wavelength converters placement problem. Both equal-cost multipath routing and traffic-engineering-aware shortest-path routing are included in the experiments to examine their effectiveness in reducing the overall blocking probability. These routing schemes help to achieve network-wide load balance to ensure higher resource utilization efficiency. Computed results show that our proposed heuristic is highly effective and efficient in reducing the blocking probability in sparse wavelength conversion networks.

Topology discovery for public IPv6 networks

In just three decades the Internet has grown from a small experimental research network into a complex network of routers, switches, and hosts. Understanding the topology of such large scale networks is essential to the procurement of good architectural design decisions, particularly with respect to address allocation and distribution schemes.A number of techniques for IPv4 network topology already exist. Of these ICMP-based probing has shown to be most useful in determining router-level topologies of public networks. However, many of these techniques cannot be readily applied to IPv6 because of changes in the addressing scheme and ICMP behaviour. Furthermore, increases in the proliferation of equal-cost multi-path routing, and other forms of transient routing, indicate that traditional traceroute-based topology discovery approaches are becoming less effective in the Internet.This paper presents Atlas, a system that faciliates the automated capture of IPv6 network topology information from a single probing host. It describes the Atlas infrastructure and its data collection processes and discusses IPv6 network phenomena that must to be taken into account by the probing scheme. We also present some initial results from our probing of the 6Bone, currently the largest public IPv6 network. The results illustrate the effectiveness of the probing algorithm and also identify some trends in prefix allocation and routing policy.