Data Center Networking

Data center networks encode locality and topology information into their server and switch addresses for performance and routing purposes. For this reason, the traditional address configuration protocols such as DHCP require huge amount of manual input, leaving them error-prone. This project introduces how to autoconfigure a large-scale data centers with millions of devices.

Data center networks (DCNs) form the backbone infrastructure of many large-scale enterprise applications as well as emerging cloud computing providers. Leveraging runtime reconfigurable optical devices, this project aim to dynamically changes its topology and link capacities, thereby achieving unprecedented flexibility to adapt to dynamic traffic patterns, with the hope to deliver high bisection bandwidth.

Recent measurement studies suggest that datacenter networks suffer from long-lived congestion caused by core network oversubscription and unbalanced workload placement. In contrast to traditional traffic engineering approaches that optimize flow routing, in this project, we explore the opportunity to address the continuous congestion via optimizing VM placement in virtualized datacenters.

Optical data center networks (DCNs) are becoming increasingly attractive due to their technological strengths compared to traditional electrical networks. However, existing optical DCNs are either hard to scale, vulnerable to single point of failure, or provide limited network bisection bandwidth for many practical DCN workloads. In this project, we design a scalable, fault-tolerant, high-performance optical DCN architecture.