RDNA Balance: Load Balancing by Isolation of Elephant Flows using Strict Source Routing

TL;DR

This paper introduces RDNA Balance, a load balancing method for data center networks that isolates large elephant flows using source routing, enhancing resource utilization with simplicity and scalability.

Contribution

It presents a novel load balancing approach leveraging elephant flow isolation and source routing, improving network resource utilization in data centers.

Findings

Effective load balancing with elephant flow isolation

Scalable and programmable solution demonstrated

Utilizes OpenFlow features for flow classification

Abstract

Data center networks need load balancing mechanisms to dynamically serve a large number of flows with different service requirements. However, traditional load-balancing approaches do not allow the full utilization of network resources in a simple, programmable, and scalable way. In this context, this paper proposes RDNA Balance that exploits elephant flow isolation and source routing in core nodes. Flow classification operations are performed on the edge using features of the OpenFlow protocol. The results show that with this approach it is possible to provide a simple, scalable, and programmable load balancing for data centers.