Reducing transport latency for short flows with multipath TCP

Dong, Pingping; Yang, Wenjun; Tang, Wensheng; Huang, Jiawei; Wang, Haodong; Pan, Yi; Wang, Jianxin

doi:10.1016/j.jnca.2018.02.005

Cited by 31 publications

(17 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Delay-based protocols (e.g., TCP Vegas and FAST TCP [23]) detect the network congestion and adjust the cwnd based on RTT. The delay-based variants can react to the network congestion more quickly compared to the loss-based mechanisms [24] and are capable of limiting the standing queue size and decreasing the amount of packets that are dropped in the eNodeB. However, the delay-based approaches will suffer from significant throughput degradation when competing with loss-based algorithms, e.g., TCP-Reno [25].…”

Section: Related Workmentioning

confidence: 99%

Receiver-Side TCP Countermeasure in Cellular Networks

Dong

Gao

Xie

et al. 2019

Sensors

Self Cite

View full text Add to dashboard Cite

Cellular-based networks keep large buffers at base stations to smooth out the bursty data traffic, which has a negative impact on the user’s Quality of Experience (QoE). With the boom of smart vehicles and phones, this has drawn growing attention. For this paper, we first conducted experiments to reveal the large delays, thus long flow completion time (FCT), caused by the large buffer in the cellular networks. Then, a receiver-side transmission control protocol (TCP) countermeasure named Delay-based Flow Control algorithm with Service Differentiation (DFCSD) was proposed to target interactive applications requiring high throughput and low delay in cellular networks by limiting the standing queue size and decreasing the amount of packets that are dropped in the eNodeB in Long Term Evolution (LTE). DFCSD stems from delay-based congestion control algorithms but works at the receiver side to avoid the performance degradation of the delay-based algorithms when competing with loss-based mechanisms. In addition, it is derived based on the TCP fluid model to maximize the network utility. Furthermore, DFCSD also takes service differentiation into consideration based on the size of competing flows to shorten their completion time, thus improving user QoE. Simulation results confirmed that DFCSD is compatible with existing TCP algorithms, significantly reduces the latency of TCP flows, and increases network throughput.

show abstract

Section: Related Workmentioning

confidence: 99%

Receiver-Side TCP Countermeasure in Cellular Networks

Dong

Gao

Xie

et al. 2019

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…In data centers, the network traffic shows the following three characters. Firstly, most flows are small in size (i.e., ≤100 KB) [1,8]. Some tiny flows are even smaller than 10 KB.…”

Section: Traffic Characters In Data Centermentioning

confidence: 99%

“…In data centers, most flows are small in size. For example, more than 80% of flows in data mining applications are usually small in length (e.g., 10 KB) [8]. It indicates that most flows are small enough 2 of 13 to finish their transmissions in the slow start phase.…”

Section: Introductionmentioning

confidence: 99%

Gentle Slow Start to Alleviate TCP Incast in Data Center Networks

2019

Self Cite

View full text Add to dashboard Cite

Modern data center networks typically adopt symmetric topologies, such as leaf-spine and fat-tree. When a large number of transmission control protocol (TCP) flows in data center networks send data to the same receiver, the congestion collapse, called TCP Incast, frequently happens because of the huge packet losses and Time-Out. To address the TCP Incast issue, we firstly demonstrate that adjusting the increasing speed of the congestion window during the slow start phase is crucially important. Then we propose the Gentle Slow Start (GSS) algorithm, which adjusts the congestion window according to real-time congestion state in a gentle manner and smoothly switches from slow start to congestion avoidance phase. Furthermore, we present the implementation and design of Gentle Slow Start and also integrate it into the state-of-the-art data center transport protocols. The test results show that GSS effectively decreases the Incast probability and increases the network goodput by average 8x.

show abstract

“…A multipath transmission control protocol (MPTCP) [20,21] has been using as a transport protocol to routes subflows toward multiple paths. This is because it cannot change paths, which is why the MPTCP routed paths congestion mismatch issue is very low [22][23][24].…”

Section: Related Workmentioning

confidence: 99%

Novel Multi-Level Dynamic Traffic Load-Balancing Protocol for Data Center

et al. 2019

Self Cite

View full text Add to dashboard Cite

Typically, the production data centers function with various risk factors, such as for instance the network dynamicity, topological asymmetry, and switch failures. Hence, the load-balancing schemes should consider the sensing accurate path circumstances as well as the reduction of failures. However, under dynamic traffic, current load-balancing schemes use the fixed parameter setting, resulting in suboptimal performances. Therefore, we propose a multi-level dynamic traffic load-balancing (MDTLB) protocol, which uses an adaptive approach of parameter setting. The simulation results show that the MDTLB outperforms the state-of-the-art schemes in terms of both the flow completion time and throughput in typical data center applications.

show abstract

Reducing transport latency for short flows with multipath TCP

Abstract: Fig. 2. The average completion time of MPTCP when the RTT of one subflow is fixed at 10 ms and the other varies from 10 ms to 200 ms.

Cited by 31 publications

References 25 publications

Receiver-Side TCP Countermeasure in Cellular Networks

Receiver-Side TCP Countermeasure in Cellular Networks

Gentle Slow Start to Alleviate TCP Incast in Data Center Networks

Novel Multi-Level Dynamic Traffic Load-Balancing Protocol for Data Center

Contact Info

Product

Resources

About