A way of managing data center networks

Abstract

Evolutionary changes have occurred throughout the various elements of the data centre, starting with server and storage virtualization and also network virtualization. Motivations for server virtualization were initially associated with massive cost reduction and redundancy but have now evolved to focus on greater scalability and agility within the data centre. Data center focused LAN technologies have taken a similar path; with a goal of redundancy and then to create a more scalable fabric within and between data centres. The TCP's congestion control opens the network to denial of service attacks and performance interference.. We propose a natural evolution of data center transport from TCP to multipath TCP. We show that multipath TCP can effectively and seamlessly use available bandwidth, providing improved throughput and better fairness in these new topologies when compared to single path TCP and randomized flow-level load balancing.