TCP congestion control macroscopic behaviour for combinations of source and router algorithms

Mulalo Dzivhani; Dumisa Ngwenya; Moshe Masonta; Khmaies Ouahada

doi:10.1109/ICASTECH.2018.8506978

TCP congestion control macroscopic behaviour for combinations of source and router algorithms

Mulalo Dzivhani, Dumisa Ngwenya, Moshe Masonta, Khmaies Ouahada

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

The network side of Transmission Control Protocol (TCP) congestion control is normally considered a black-box in performance analysis. However, the overall performance of TCP/IP networks is affected by selection of congestion control mechanisms implemented at the source nodes as well as those implemented at the routers. The paper presents an evaluation of macroscopic behaviour of TCP for various combinations of source algorithms and router algorithms using a Dumbbell topology. In particular, we are interested in the throughput and fairness index. TCP New Reno and TCP Cubic were selected for source nodes. Packet First-in-First-out (PFIFO) and Controlled Delay (CoDel) mechanisms were selected for routers. The results show that TCP New Reno performs well, in terms of throughput, in a low BDP scenario. However, as expected in high BDP scenario, TCP New Reno deteriorates and TCP Cubic performs better. CoDel in the network side further deteriorates TCP New Reno flows in high Bandwidth-Delay Product (BDP) scenario, while considerably improving TCP Cubic. PFIFO deteriorates both TCP Cubic and TCP New Reno in high BDP. Almost in all cases CoDel seems to improve fairness.

Original language	English
Title of host publication	2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018
Publisher	IEEE Computer Society
ISBN (Electronic)	9781538642337
DOIs	https://doi.org/10.1109/ICASTECH.2018.8506978
Publication status	Published - 24 Oct 2018
Externally published	Yes
Event	7th IEEE International Conference on Adaptive Science and Technology, ICAST 2018 - Accra, Ghana Duration: 22 Aug 2018 → 24 Aug 2018

Publication series

Name	IEEE International Conference on Adaptive Science and Technology, ICAST
Volume	2018-August
ISSN (Print)	2326-9413
ISSN (Electronic)	2326-9448

Conference

Conference	7th IEEE International Conference on Adaptive Science and Technology, ICAST 2018
Country/Territory	Ghana
City	Accra
Period	22/08/18 → 24/08/18

Keywords

CoDel
Droptail
Fairness
TCP Congestion Control
TCP Cubic
TCP New Reno
Throughput

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems
Software
Electrical and Electronic Engineering
Communication

Access to Document

10.1109/ICASTECH.2018.8506978

Cite this

Dzivhani, M., Ngwenya, D., Masonta, M., & Ouahada, K. (2018). TCP congestion control macroscopic behaviour for combinations of source and router algorithms. In 2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018 Article 8506978 (IEEE International Conference on Adaptive Science and Technology, ICAST; Vol. 2018-August). IEEE Computer Society. https://doi.org/10.1109/ICASTECH.2018.8506978

@inproceedings{262d8214fe0342e2ab2c9c8fbca5ab01,

title = "TCP congestion control macroscopic behaviour for combinations of source and router algorithms",

abstract = "The network side of Transmission Control Protocol (TCP) congestion control is normally considered a black-box in performance analysis. However, the overall performance of TCP/IP networks is affected by selection of congestion control mechanisms implemented at the source nodes as well as those implemented at the routers. The paper presents an evaluation of macroscopic behaviour of TCP for various combinations of source algorithms and router algorithms using a Dumbbell topology. In particular, we are interested in the throughput and fairness index. TCP New Reno and TCP Cubic were selected for source nodes. Packet First-in-First-out (PFIFO) and Controlled Delay (CoDel) mechanisms were selected for routers. The results show that TCP New Reno performs well, in terms of throughput, in a low BDP scenario. However, as expected in high BDP scenario, TCP New Reno deteriorates and TCP Cubic performs better. CoDel in the network side further deteriorates TCP New Reno flows in high Bandwidth-Delay Product (BDP) scenario, while considerably improving TCP Cubic. PFIFO deteriorates both TCP Cubic and TCP New Reno in high BDP. Almost in all cases CoDel seems to improve fairness.",

keywords = "CoDel, Droptail, Fairness, TCP Congestion Control, TCP Cubic, TCP New Reno, Throughput",

author = "Mulalo Dzivhani and Dumisa Ngwenya and Moshe Masonta and Khmaies Ouahada",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 7th IEEE International Conference on Adaptive Science and Technology, ICAST 2018 ; Conference date: 22-08-2018 Through 24-08-2018",

year = "2018",

month = oct,

day = "24",

doi = "10.1109/ICASTECH.2018.8506978",

language = "English",

series = "IEEE International Conference on Adaptive Science and Technology, ICAST",

publisher = "IEEE Computer Society",

booktitle = "2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018",

address = "United States",

}

Dzivhani, M, Ngwenya, D, Masonta, M & Ouahada, K 2018, TCP congestion control macroscopic behaviour for combinations of source and router algorithms. in 2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018., 8506978, IEEE International Conference on Adaptive Science and Technology, ICAST, vol. 2018-August, IEEE Computer Society, 7th IEEE International Conference on Adaptive Science and Technology, ICAST 2018, Accra, Ghana, 22/08/18. https://doi.org/10.1109/ICASTECH.2018.8506978

TCP congestion control macroscopic behaviour for combinations of source and router algorithms. / Dzivhani, Mulalo; Ngwenya, Dumisa; Masonta, Moshe et al.
2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018. IEEE Computer Society, 2018. 8506978 (IEEE International Conference on Adaptive Science and Technology, ICAST; Vol. 2018-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - TCP congestion control macroscopic behaviour for combinations of source and router algorithms

AU - Dzivhani, Mulalo

AU - Ngwenya, Dumisa

AU - Masonta, Moshe

AU - Ouahada, Khmaies

PY - 2018/10/24

Y1 - 2018/10/24

N2 - The network side of Transmission Control Protocol (TCP) congestion control is normally considered a black-box in performance analysis. However, the overall performance of TCP/IP networks is affected by selection of congestion control mechanisms implemented at the source nodes as well as those implemented at the routers. The paper presents an evaluation of macroscopic behaviour of TCP for various combinations of source algorithms and router algorithms using a Dumbbell topology. In particular, we are interested in the throughput and fairness index. TCP New Reno and TCP Cubic were selected for source nodes. Packet First-in-First-out (PFIFO) and Controlled Delay (CoDel) mechanisms were selected for routers. The results show that TCP New Reno performs well, in terms of throughput, in a low BDP scenario. However, as expected in high BDP scenario, TCP New Reno deteriorates and TCP Cubic performs better. CoDel in the network side further deteriorates TCP New Reno flows in high Bandwidth-Delay Product (BDP) scenario, while considerably improving TCP Cubic. PFIFO deteriorates both TCP Cubic and TCP New Reno in high BDP. Almost in all cases CoDel seems to improve fairness.

AB - The network side of Transmission Control Protocol (TCP) congestion control is normally considered a black-box in performance analysis. However, the overall performance of TCP/IP networks is affected by selection of congestion control mechanisms implemented at the source nodes as well as those implemented at the routers. The paper presents an evaluation of macroscopic behaviour of TCP for various combinations of source algorithms and router algorithms using a Dumbbell topology. In particular, we are interested in the throughput and fairness index. TCP New Reno and TCP Cubic were selected for source nodes. Packet First-in-First-out (PFIFO) and Controlled Delay (CoDel) mechanisms were selected for routers. The results show that TCP New Reno performs well, in terms of throughput, in a low BDP scenario. However, as expected in high BDP scenario, TCP New Reno deteriorates and TCP Cubic performs better. CoDel in the network side further deteriorates TCP New Reno flows in high Bandwidth-Delay Product (BDP) scenario, while considerably improving TCP Cubic. PFIFO deteriorates both TCP Cubic and TCP New Reno in high BDP. Almost in all cases CoDel seems to improve fairness.

KW - CoDel

KW - Droptail

KW - Fairness

KW - TCP Congestion Control

KW - TCP Cubic

KW - TCP New Reno

KW - Throughput

UR - http://www.scopus.com/inward/record.url?scp=85056846108&partnerID=8YFLogxK

U2 - 10.1109/ICASTECH.2018.8506978

DO - 10.1109/ICASTECH.2018.8506978

M3 - Conference contribution

AN - SCOPUS:85056846108

T3 - IEEE International Conference on Adaptive Science and Technology, ICAST

BT - 2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018

PB - IEEE Computer Society

T2 - 7th IEEE International Conference on Adaptive Science and Technology, ICAST 2018

Y2 - 22 August 2018 through 24 August 2018

ER -

Dzivhani M, Ngwenya D, Masonta M, Ouahada K. TCP congestion control macroscopic behaviour for combinations of source and router algorithms. In 2018 IEEE 7th International Conference on Adaptive Science and Technology, ICAST 2018. IEEE Computer Society. 2018. 8506978. (IEEE International Conference on Adaptive Science and Technology, ICAST). doi: 10.1109/ICASTECH.2018.8506978

TCP congestion control macroscopic behaviour for combinations of source and router algorithms

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this