Integration and Evaluation of QUIC and TCP-BBR in longhaul Science Data Transfers
Jisc, Lumen House, Library Avenue, Harwell Campus,
Oxfordshire OX11 0SG,
2 College of Engineering and Technology, University of Derby, Derby, DE22 1GB, UK
3 College of Engineering, Design and Physical Sciences, Brunel University London, Uxbridge, UB8 3PH, UK
* e-mail: email@example.com
** e-mail: firstname.lastname@example.org
*** e-mail: email@example.com
Published online: 17 September 2019
Two recent and promising additions to the internet protocols are TCP-BBR and QUIC. BBR defines a congestion policy that promises a better control in TCP bottlenecks on long haul transfers and can also be used in the QUIC protocol. TCP-BBR is implemented in the Linux kernels above 4.9. It has been shown, however, to demand careful fine tuning in the interaction, for example, with the Linux Fair Queue. QUIC, on the other hand, replaces HTTP and TLS with a protocol on the top of UDP and thin layer to serve HTTP. It has been reported to account today for 7% of Google’s traffic. It has not been used in server-to-server transfers even if its creators see that as a real possibility.
Our work evaluates the applicability and tuning of TCP-BBR and QUIC for data science transfers. We describe the deployment and performance evaluation of TCP-BBR and comparison with CUBIC and H-TCP in transfers through the TEIN link to Singaren (Singapore). Also described is the deployment and initial evaluation of a QUIC server.
We argue that QUIC might be a perfect match in security and connectivity to base services that are today performed by the Xroot redirectors.
© The Authors, published by EDP Sciences, 2019
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.