A First Look at Real Multipath TCP Traffic

Multi-path TCP (MP-TCP) has the prospective to significantly advance applications performance by using multiple paths evidently. Multipath TCP was intended and employed as a backward compatible replacement for TCP. For this reason, it exposes the standard socket API to the applications that cannot control the utilization of the different paths. This is a key feature for applications that are unaware of the multipath nature of the network. On the contrary, this is a limitation for applications that could benefit from specific knowledge to use multiple paths in a way that fits their needs. Therefore, hosts are often connected by multiple paths, but TCP restricts communications to a single path per transport connection. Resource usage within the network would be more efficient where these multiple paths able to be used concurrently. This should enhance user experience through improved resilience to network failure and higher throughput. In this paper, we have focused on MPTCP and discussed the performance issues and its solution. We believe our concept will be useful for future works of MPTCP performance evaluation

Proceedings of the 2016 workshop on Applied Networking Research Workshop - ANRW 16, 2016

Multipath TCP has been recently introduced in order to allow a better resource consumption and user quality-ofexperience. This is achieved by allowing a connection between two hosts through multiple subflows. However, with the rise of middleboxes and inherent Internet ossification, the large-scale deployment of this TCP extension is difficult. In particular, a load balancer at the entry point of a data center may forward subflows to different servers, cancelling so the advantages of Multipath TCP.

Proceedings of the 9th ACM SIGCOMM conference on Internet measurement, 2009

Since the last in-depth studies of measured TCP traffic some 6-8 years ago, the Internet has experienced significant changes, including the rapid deployment of backbone links with 1-2 orders of magnitude more capacity, the emergence of bandwidth-intensive streaming applications, and the massive penetration of new TCP variants. These and other changes beg the question whether the characteristics of measured TCP traffic in today's Internet reflect these changes or have largely remained the same. To answer this question, we collected and analyzed packet traces from a number of Internet backbone and access links, focused on the "heavy-hitter" flows responsible for the majority of traffic. Next we analyzed their within-flow packet dynamics, and observed the following features: (1) in one of our datasets, up to 15.8% of flows have an initial congestion window (ICW) size larger than the upper bound specified by RFC 3390. (2) Among flows that encounter retransmission rates of more than 10%, 5% of them exhibit irregular retransmission behavior where the sender does not slow down its sending rate during retransmissions. (3) TCP flow clocking (i.e., regular spacing between flights of packets) can be caused by both RTT and non-RTT factors such as application or link layer, and 60% of flows studied show no pronounced flow clocking. To arrive at these findings, we developed novel techniques for analyzing unidirectional TCP flows, including a technique for inferring ICW size, a method for detecting irregular retransmissions, and a new approach for accurately extracting flow clocks.

2017

The evolution in networking coupled with an increasing demand to improve user experience has led to different proposals to extend the standard TCP. Multipath TCP (MPTCP) is one such extension that ...

… Infrastructure for the …

We present an implementation of Multipath TCP (MPTCP) under the NS-3 open source network simulator. MPTCP is a promising extension of TCP currently considered by the recent eponymous IETF working group, with the objective of improving the performance of TCP, especially its robustness to variable network conditions. We describe this new protocol, its main functions and our implementation in NS-3. Besides this implementation compliant to the current versions of the IETF drafts, we have also added and compared various packet reordering mechanisms. We indeed notice that such mechanisms highly improve the performance of MPTCP. We believe that our implementation could be useful for future works in MPTCP performance evaluation, especially to compare packet reordering algorithms or coupling congestion control mechanisms between subflows.

Proceedings of the …, 2005

International Conference on Broadband Networks, 2009

Multi-interface mobile devices and multihomed residential Internet connections are becoming commonplace. However, standard transport protocols TCP and SCTP are unable to take advantage of several available paths so that the application using a single transport connection would receive the aggregate bandwidth of all paths. Multihoming and advanced security features make the Host Identity Protocol a good candidate to provide multipath data delivery. In this paper, we design and implement a multipath scheduler that distributes the incoming traffic among multiple available paths. Using Fastest Path First scheduling, packets from a single TCP connection could be spread to multiple paths with no reordering. Our simulations confirm effectiveness and TCP-friendliness of multipath transfer for a range of path bandwidths and in the presence of cross-traffic. 1

IEEE Access, 2019

Ethernet switched networks are widely used in enterprise and data center networks. However, they have some drawbacks, mainly that, to prevent loops, they cannot take advantage of multipath topologies to balance traffic. Several multipath routing proposals use link-state protocols and Equal Cost Multi-Path routing (ECMP) to distribute the load over multiple paths. But, these proposals are complex and prone to flow collisions that may degrade performance. This paper studies TCP-Path, a protocol that employs a different approach. It uses a distributed network exploration mechanism based on broadcasting the TCP-SYN packet to identify and select the fastest available path to the destination host, on the fly. Our evaluation shows that it improves on ECMP by up to 70% in terms of throughput for elephant flows and by up to 60% in terms of flow completion time for mouse flows. Indeed, network exploration offers a better, yet simple alternative to ECMP-based solutions for multipath topologies. In addition, we also study TCP-Path for elephant flows (TFE), which restricts TCP-Path application to elephant flows to reduce the exploration broadcast overhead and the size of forwarding tables, thus improving its scalability. Although elephant flows represent a small fraction (about 5%) of total flows, they have a major impact on overall performance, as we show in our evaluation. TFE reduces both the overhead incurred during path setup and the size of the forwarding tables by a factor of almost 20. Moreover, it achieves results close to those obtained by TCP-Path for elephant flows, especially when working with high loads, and yields significant improvements for all types of flow at medium and high load levels.

ACM SIGCOMM Computer Communication Review, 2011

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.