Skip to main content

Path Inference in Data Center Networks

Researcher: Hyong Kim

Research Area: Next Generation Secure and Available Networks

Abstract

Scope: Data center networks have started to play increasingly important roles in today’s Internet. Popular web-based services and critical enterprise applications are hosted in large data centers. More recent advances like cloud computing and cellular-based data usage have only increased the importance of data centers. With the increasing importance, however, also comes increasing complexity. Supporting the wide-array of applications and traffic types while meeting all their performance and security requirements results in complex network designs. The result of this complexity is that managing these networks has never been more difficult. In this project, we focus on providing one of the key building blocks of network management: the ability to determine how traffic flows in the network. This information is fundamental to many different network management tasks including troubleshooting, capacity planning, and what-if analysis. Towards that end, we develop a system that performs per-packet path inference in a data center. The proposed system uses device configurations and the network’s physical topology to project the path of a packet in the network. Specifically, per-hop path inference based on a simplified model of layer-3 routing, layer-2 switching, and the most commonly used routing and forwarding mechanisms. To show its applicability, we perform path inference within a campus network and on multiple data centers serving the 3G network of a major cellular service provider. Using routing information collected from these networks, we validate the correctness of the inferred paths. Our system is expected to quickly and accurately determine paths traversed by packets even in complex data center networks, making it a valuable addition to a network operator’s toolbox.