Improving datacenter performance and robustness with multipath TCP
Sun, 05/15/2011 - 19:37 by Sébastien Barré
Abstract
The latest large-scale data centers offer higher aggregate bandwidth and robustness by creating multiple paths in the core of the network. To utilize this bandwidth requires different flows take different paths, which poses a challenge. In short, a single-path transport seems ill-suited to such networks. We propose using Multipath TCP as a replacement for TCP in such data centers, as it can effectively and seamlessly use available bandwidth, giving improved throughput and better fairness on many topologies. We investigate the reasons behind these benefits, teasing apart the contribution of each of the mechanisms used by MPTCP.
Using MPTCP allows us to rethink data center networks, with a different mindset as to the relationship between transport protocols, routing and topology. MPTCP enables better topologies that single path TCP just can't use. As a proof-of-concept, we present a dual-homed variant of the FatTree topology. Using MPTCP, this outperforms FatTree for a wide range of workloads, but costs the
same.
In existing data center networks, MPTCP is readily deployable as it can leverage widely deployed technologies such as ECMP. We have run MPTCP on Amazon EC2 and found that it commonly outperforms TCP by a factor of three. But the biggest benefits will come when data centers are designed around the use of multipath transports.
- Authors
- C. Raiciu, S. Barré, C. Pluntke, A. Greenhalgh, D. Wischik and M. Handley
- Source
SIGCOMM 2011, Toronto, Canada , August 2011.- Notes
- See http://www.multipath-tcp.org for related work on MPTCP and the Linux kernel implementation used in this paper
- Full text
-
(434.68 KB) - Cite it
- BibTex
- Copyright
- See here
IEEE Copyright Notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.
ACM Copyright Notice: Copyright 1999 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page or intial screen of the document. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept., ACM Inc., fax +1 (212) 869-0481, or permissions@acm.org.
Springer-Verlag LNCS Copyright Notice: The copyright of these contributions has been transferred to Springer-Verlag Berlin Heidelberg New York. The copyright transfer covers the exclusive right to reproduce and distribute the contribution, including reprints, translations, photographic reproductions, microform, electronic form (offline, online), or any other reproductions of similar nature. Online available from Springer-Verlag LNCS series.