Measuring and extending Multipath TCP

Mon, 05/25/2020 - 13:03 by Olivier Bonaventure


TCP has been one of the most important Internet protocols since the early days of this network. The initial version of TCP assumed that (1) each device has a single interface, and (2) its network address is permanent. Today's Internet attached devices have multiple interfaces with dynamic addresses. These deployments do not match anymore the design principles of TCP. By decoupling the transport layer from the underneath IP layer, Multipath TCP brings several key benefits in a variety of use cases. However, this major TCP extension is also significantly more complex than the legacy TCP. Despite growing interests in Multipath TCP, there are still many unknowns about its behaviours and performance in the real world. Moreover, most Multipath TCP implementations are based on existing TCP stacks which are part of operating systems kernels. Therefore, it is challenging to build Multipath TCP stacks that adapt to different network scenarios and user requirements. The purpose of this thesis is to answer two main research questions: (1) how Multipath TCP and its current implementations behave and perform in the Internet, and (2) how to customize and extend Multipath TCP implementations to adapt them to current and future use cases. For the first question, we have conducted two measurement campaigns, one with traditional web services and the other with voice-activated services. For the second question, we have explored the capabilities of the eBPF infrastructure in the Linux kernel and leveraged it to extend both the TCP and the Multipath TCP stacks in the Linux kernel.

Viet-Hoang Tran
PhD thesis
UCLouvain, 2019.
Full text
pdf   (3.57 MB)
Cite it
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

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.