Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols

Ayan Shymyrbay, Arshyn Zhanbolatov, Assilkhan Amankhan, Adilya Bakambekova, Ikechi Augustine Ukaegbu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Data centers had come to the scene as an effective framework for running a huge range of real-time and cloud applications. When TCP is deployed in a broad data center environment, it fails to maintain high throughput and deadline guarantees, mainly due to the unfair sharing and flow quenching. Several transport layer designs have been proposed over the last few years: some of them were claimed to outperform the currently used ones. This paper provides an analysis of some of the transport-layer protocols, namely D 3 Deadline Driven Delivery), PDQ (Preemptive Distributed Quick), D 2 TCP (Deadline-aware Data Center TCP), DCTCP (Data Center TCP) and MCP (Multiflow Conversation Protocol). The analysis includes the evaluation of their design implementation, performance and problems. The FCFS mechanisms at the heart of the D 3 forces the protocol to be tied to the arrival order of the flows, which leads to the priority inversion issue resulting in flows unjustifiably missing their deadlines. The PDQ protocol has been described as a mechanism for a flow urgency regulation which emulates the EDF behavior. While PDQ has improved upon D 3 , it was pointed out that the management of priority lists and per-flow states in this line of protocols puts a great burden on the network core, thus restricting the efficient router's bandwidth. Motivated by DTCP, D 2 TCP and MCP utilize congestion window modeling at the end hosts to handle the deadline flows ensuring network stability and deadline meeting guarantees. Also, deployability problem is likely to be solved by recent advances of SDN introducing a centralized control over networks.
Original languageEnglish
Title of host publication2018 International Conference on Computing and Network Communications (CoCoNet)
PublisherIEEE
Pages152-158
Number of pages7
Publication statusPublished - Oct 1 2018

Fingerprint

Electric network analysis
Network protocols
Routers
Quenching
Throughput
Bandwidth

Cite this

Shymyrbay, A., Zhanbolatov, A., Amankhan, A., Bakambekova, A., & Ukaegbu, I. A. (2018). Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols. In 2018 International Conference on Computing and Network Communications (CoCoNet) (pp. 152-158). IEEE.

Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols. / Shymyrbay, Ayan; Zhanbolatov, Arshyn; Amankhan, Assilkhan; Bakambekova, Adilya; Ukaegbu, Ikechi Augustine.

2018 International Conference on Computing and Network Communications (CoCoNet). IEEE, 2018. p. 152-158.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shymyrbay, A, Zhanbolatov, A, Amankhan, A, Bakambekova, A & Ukaegbu, IA 2018, Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols. in 2018 International Conference on Computing and Network Communications (CoCoNet). IEEE, pp. 152-158.
Shymyrbay A, Zhanbolatov A, Amankhan A, Bakambekova A, Ukaegbu IA. Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols. In 2018 International Conference on Computing and Network Communications (CoCoNet). IEEE. 2018. p. 152-158
Shymyrbay, Ayan ; Zhanbolatov, Arshyn ; Amankhan, Assilkhan ; Bakambekova, Adilya ; Ukaegbu, Ikechi Augustine. / Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols. 2018 International Conference on Computing and Network Communications (CoCoNet). IEEE, 2018. pp. 152-158
@inproceedings{bac7be795ce24da68124c2be3b77afa1,
title = "Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols",
abstract = "Data centers had come to the scene as an effective framework for running a huge range of real-time and cloud applications. When TCP is deployed in a broad data center environment, it fails to maintain high throughput and deadline guarantees, mainly due to the unfair sharing and flow quenching. Several transport layer designs have been proposed over the last few years: some of them were claimed to outperform the currently used ones. This paper provides an analysis of some of the transport-layer protocols, namely D 3 Deadline Driven Delivery), PDQ (Preemptive Distributed Quick), D 2 TCP (Deadline-aware Data Center TCP), DCTCP (Data Center TCP) and MCP (Multiflow Conversation Protocol). The analysis includes the evaluation of their design implementation, performance and problems. The FCFS mechanisms at the heart of the D 3 forces the protocol to be tied to the arrival order of the flows, which leads to the priority inversion issue resulting in flows unjustifiably missing their deadlines. The PDQ protocol has been described as a mechanism for a flow urgency regulation which emulates the EDF behavior. While PDQ has improved upon D 3 , it was pointed out that the management of priority lists and per-flow states in this line of protocols puts a great burden on the network core, thus restricting the efficient router's bandwidth. Motivated by DTCP, D 2 TCP and MCP utilize congestion window modeling at the end hosts to handle the deadline flows ensuring network stability and deadline meeting guarantees. Also, deployability problem is likely to be solved by recent advances of SDN introducing a centralized control over networks.",
author = "Ayan Shymyrbay and Arshyn Zhanbolatov and Assilkhan Amankhan and Adilya Bakambekova and Ukaegbu, {Ikechi Augustine}",
year = "2018",
month = "10",
day = "1",
language = "English",
pages = "152--158",
booktitle = "2018 International Conference on Computing and Network Communications (CoCoNet)",
publisher = "IEEE",

}

TY - GEN

T1 - Meeting Deadlines in Datacenter Networks: An Analysis on Deadline-Aware Transport Layer Protocols

AU - Shymyrbay, Ayan

AU - Zhanbolatov, Arshyn

AU - Amankhan, Assilkhan

AU - Bakambekova, Adilya

AU - Ukaegbu, Ikechi Augustine

PY - 2018/10/1

Y1 - 2018/10/1

N2 - Data centers had come to the scene as an effective framework for running a huge range of real-time and cloud applications. When TCP is deployed in a broad data center environment, it fails to maintain high throughput and deadline guarantees, mainly due to the unfair sharing and flow quenching. Several transport layer designs have been proposed over the last few years: some of them were claimed to outperform the currently used ones. This paper provides an analysis of some of the transport-layer protocols, namely D 3 Deadline Driven Delivery), PDQ (Preemptive Distributed Quick), D 2 TCP (Deadline-aware Data Center TCP), DCTCP (Data Center TCP) and MCP (Multiflow Conversation Protocol). The analysis includes the evaluation of their design implementation, performance and problems. The FCFS mechanisms at the heart of the D 3 forces the protocol to be tied to the arrival order of the flows, which leads to the priority inversion issue resulting in flows unjustifiably missing their deadlines. The PDQ protocol has been described as a mechanism for a flow urgency regulation which emulates the EDF behavior. While PDQ has improved upon D 3 , it was pointed out that the management of priority lists and per-flow states in this line of protocols puts a great burden on the network core, thus restricting the efficient router's bandwidth. Motivated by DTCP, D 2 TCP and MCP utilize congestion window modeling at the end hosts to handle the deadline flows ensuring network stability and deadline meeting guarantees. Also, deployability problem is likely to be solved by recent advances of SDN introducing a centralized control over networks.

AB - Data centers had come to the scene as an effective framework for running a huge range of real-time and cloud applications. When TCP is deployed in a broad data center environment, it fails to maintain high throughput and deadline guarantees, mainly due to the unfair sharing and flow quenching. Several transport layer designs have been proposed over the last few years: some of them were claimed to outperform the currently used ones. This paper provides an analysis of some of the transport-layer protocols, namely D 3 Deadline Driven Delivery), PDQ (Preemptive Distributed Quick), D 2 TCP (Deadline-aware Data Center TCP), DCTCP (Data Center TCP) and MCP (Multiflow Conversation Protocol). The analysis includes the evaluation of their design implementation, performance and problems. The FCFS mechanisms at the heart of the D 3 forces the protocol to be tied to the arrival order of the flows, which leads to the priority inversion issue resulting in flows unjustifiably missing their deadlines. The PDQ protocol has been described as a mechanism for a flow urgency regulation which emulates the EDF behavior. While PDQ has improved upon D 3 , it was pointed out that the management of priority lists and per-flow states in this line of protocols puts a great burden on the network core, thus restricting the efficient router's bandwidth. Motivated by DTCP, D 2 TCP and MCP utilize congestion window modeling at the end hosts to handle the deadline flows ensuring network stability and deadline meeting guarantees. Also, deployability problem is likely to be solved by recent advances of SDN introducing a centralized control over networks.

M3 - Conference contribution

SP - 152

EP - 158

BT - 2018 International Conference on Computing and Network Communications (CoCoNet)

PB - IEEE

ER -