Towards a Cognitive Routing Engine for Software Defined Networks

TL;DR

This paper introduces the Cognitive Routing Engine (CRE), an SDN application that efficiently finds near-optimal QoS paths with significantly reduced monitoring overhead, improving response times and network performance.

Contribution

The paper presents the design, architecture, and initial evaluation of CRE, demonstrating its ability to reduce monitoring overhead while maintaining near-optimal routing in SDN.

Findings

CRE finds near-optimal paths with 1.65% optimality gap.

CRE uses 9.5 times less monitoring than global methods.

Initial evaluation on GEANT network shows promising results.

Abstract

Most Software Defined Networks (SDN) traffic engineering applications use excessive and frequent global monitoring in order to find the optimal Quality-of-Service (QoS) paths for the current state of the network. In this work, we present the motivations, architecture and initial evaluation of a SDN application called Cognitive Routing Engine (CRE) which is able to find near-optimal paths for a user-specified QoS while using a very small monitoring overhead compared to global monitoring which is required to guarantee that optimal paths are found. Smaller monitoring overheads bring the advantage of smaller response time for the SDN controllers and switches. The initial evaluation of CRE on a SDN representation of the GEANT academic network shows that it is possible to find near-optimal paths with a small optimality gap of 1.65% while using 9.5 times less monitoring.