Applying TCP-Friendly Congestion Control to Concurrent Multipath Transfer

Dreibholz, Thomas; Becke, Martin; Pulinthanath, Jobin; Rathgeb, Erwin P.

doi:10.1109/aina.2010.117

Cited by 24 publications

(18 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This problem has been addressed by several researchers and coupled congestion controllers have been proposed. Examples include e.g., CMT/RPv1 [29] and CMT/RPv2 [30]. The problem of not considering shared bottlenecks was addressed already in the design phase of MPTCP.…”

Section: Congestion Control For Multi-path Trans-mentioning

confidence: 99%

Is multi-path transport suitable for latency sensitive traffic?

et al. 2016

Self Cite

View full text Add to dashboard Cite

This paper assesses whether multi-path communication can help latency-sensitive applications to satisfy the requirements of their users. We consider Concurrent Multi-path Transfer for SCTP (CMT-SCTP) and Multi-path TCP (MPTCP) and evaluate their proficiency in transporting video, gaming, and web traffic over combinations of WLAN and 3G interfaces. To ensure the validity of our evaluation, several experimental approaches were used including simulation, emulation and live experiments. When paths are symmetric in terms of capacity, delay and loss rate, we find that the experienced latency is significantly reduced, compared to using a single path. Using multiple asymmetric paths does not affect latencyapplications do not experience any increase or decrease, but might benefit from other advantages of multipath communication. In the light of our conclusions, multi-path transport is suitable for latency-sensitive traffic and mature enough to be widely deployed.

show abstract

Section: Congestion Control For Multi-path Trans-mentioning

confidence: 99%

Is multi-path transport suitable for latency sensitive traffic?

et al. 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…The value of β is a tradeoff, with false positives-where noise in delay signals is incorrectly deemed congestion-if too high, and with false negatives-where congestion-based delay signals are missedif too low. β = 0.3 has been found to perform well over a range of scenarios in our simulations, and is the value that we use in our evaluations 3 . The threshold R th (k + 1) is then used to detect the next delay congestion signal.…”

Section: B Congestion and Shared Bottleneck Detectionmentioning

confidence: 95%

“…In this state, the reception of new ACKs will result in an update of the congestion windows W r and W Si as per (3). That is, as long as no congestion is detected while in this state, the subflow window will increase as if in congestion avoidance.…”

Section: ) Fr/fr Without Window Reductionmentioning

confidence: 97%

Dynamic Window Coupling for multipath congestion control

Hassayoun

Iyengar

Ros

2011

2011 19th IEEE International Conference on Network Protocols

View full text Add to dashboard Cite

The traditional problem of end-hosts efficiently and fairly utilizing end-to-end paths becomes significantly harder when the end-hosts are multihomed. Such is the case, for instance, when an end-host has simultaneous connectivity through several service providers, or when a mobile device is simultaneously connected via both a wireless LAN and a cellular network. A multihoming-aware transport protocol, such as MPTCP or SCTP, that sends data over the multiple resulting end-to-end paths must be fair to other flows in the network while being able to maximize its own throughput. In this paper, we present Dynamic Window Coupling (DWC), a multipath congestion control mechanism that seeks to achieve both these goals. DWC uses loss and delay signals to detect shared bottlenecks, explicitly grouping and sharing congestion control across subflows on paths that have a common bottleneck, while separating congestion control for subflows on paths with distinct bottlenecks. DWC detects shifting bottlenecks in the network and responds by dynamically regrouping subflows. Simulations demonstrate that DWC detects shared bottlenecks under most network topologies and conditions that we considered, regroups subflows correctly as bottlenecks shift, aggregates throughput across distinct bottlenecks, and is fair to other TCP flows at all bottlenecks.

show abstract

“…In this simulation study, the load sharing of SCTP aggregated the bandwidths of all active transmission paths among connected terminals. Dreibholz et al (2010Dreibholz et al ( , 2011aDreibholz et al ( , 2011bDreibholz et al ( , 2012 combined CMT with resource pool notation to improve non-CMT transfer and presented new open-source tools to measure and evaluate both CMT-SCTP and multipath TCP. Similarly, Arshad and Saleem (2010) investigated the end-to-end transport layer protocol delay and compared FAST TCP and SCTP through simulation.…”

Section: Related Workmentioning

confidence: 98%