Fairness Improvement of Different TCP Congestion Control Algorithms Using P4 Programmable Data Plane

A wide variety of congestion control algorithms have been developed for the transport protocol of TCP/IP. A related study revealed that the communication performance becomes unfair depending on the buffer length of switches or routers in the end-to-end path when communication flows using CUBIC TCP and Bottleneck Bandwidth and Round-trip propagation time (BBR) compete with each other. Therefore, we propose the fairness improvement method between different types of congestion control in the programmable network by using P4 programming. The proposed method provides multiple queues to each congestion control in the switches or the routers for relaying. The proposed method then assigns incoming TCP flows to those queues for each congestion control algorithm. The congestion control algorithm is determined by P4, which implements a programmable data plane. Three types of packet scheduling are used to ensure fairness when sending packets. We evaluate the effectiveness of the proposed method using the network emulator “mininet” and the communication performance when the queue length of the relaying switch.

Authors

Institutions

Publication Details

Published
2024-08-21
DOI
https://doi.org/10.1109/pacrim61180.2024.10690198
Citations
3
Primary Topic
Network Traffic and Congestion Control
Type
article
Field-Weighted Citation Impact
1.05
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Fairness Improvement of Different TCP Congestion Control Algorithms Using P4 Programmable Data Plane

Daiki Nobayashi, Takeshi Ikenaga, Koshi Ono
3 citations
Network Traffic and Congestion Control
1.05
article

Fairness Improvement of Different TCP Congestion Control Algorithms Using P4 Programmable Data Plane

Daiki Nobayashi, Takeshi Ikenaga, Koshi Ono
article en
3 citations

Abstract

A wide variety of congestion control algorithms have been developed for the transport protocol of TCP/IP. A related study revealed that the communication performance becomes unfair depending on the buffer length of switches or routers in the end-to-end path when communication flows using CUBIC TCP and Bottleneck Bandwidth and Round-trip propagation time (BBR) compete with each other. Therefore, we propose the fairness improvement method between different types of congestion control in the programmable network by using P4 programming. The proposed method provides multiple queues to each congestion control in the switches or the routers for relaying. The proposed method then assigns incoming TCP flows to those queues for each congestion control algorithm. The congestion control algorithm is determined by P4, which implements a programmable data plane. Three types of packet scheduling are used to ensure fairness when sending packets. We evaluate the effectiveness of the proposed method using the network emulator “mininet” and the communication performance when the queue length of the relaying switch.

Kyushu Institute of Technology (JP)
Openalex Percentile: Top 21%
Network Traffic and Congestion Control
1.05
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Fairness Improvement of Different TCP Congestion Control Algorithms Using P4 Programmable Data Plane — Daiki Nobayashi, Takeshi Ikenaga, et al. · (2024) | TGRS Research Map | TGRS