RAPID: Avoiding TCP Incast Throughput Collapse in Public Clouds with Intelligent Packet Discarding

Yang Xu, Shikhar Shukla, Zehua Guo, Sen Liu, Adrian S.W. Tam, Kang Xi, H. Jonathan Chao

Research output: Contribution to journalArticle

Abstract

Many applications in public clouds require a high fan-in, many-to-one type of data communication (known as TCP incast) in modern Data Center Networks (DCNs). Such communication could cause severe incast congestion in switches and result in TCP throughput collapse, substantially degrading the application performance. The root cause of throughput collapse is the Retransmission Timeouts (RTO) due to packet losses in congested switches. Tenants in public clouds can opt to use a variety of TCP versions. However, the existing solutions rely on modifications of TCP protocols and specific techniques from switches, and thus these existing solutions are not always feasible for public clouds. In this paper, we are inspired by the emerging virtualization and network softwarization technologies to develop a novel scheme called Retransmission timeout Avoidance by Packet Intelligent Discarding (RAPID) using software switches. RAPID considers the number of packets of each incast flow, buffered in the switch to selectively discard some packets, and ensures that the Fast Retransmission/Fast Recovery rather than RTO is invoked at the sender(s) in response to packet loss. Thus, the long idle period of a timeout and the throughput drop are avoided. We prove that, given a predetermined minimum switch buffer space, dedicated to the incast application, RAPID can prevent RTO in all the incast senders. We also present a low-complexity heuristic version of RAPID named RAPID-ED, which combines the principles of RAPID and early detection and is extremely easy to implement on today's software switches. We evaluate the two proposed schemes in a data center network testbed built on NS-3 simulator. The simulation results confirm the theoretical expectation, and show that the RAPID and RAPID-ED perform very well to prevent RTO of TCP incast flows and hence the throughput collapse. Compared with other incast solutions, RAPID and RAPID-ED do not modify TCP protocols and therefore are more suitable in public clouds.

Original languageEnglish (US)
Article number8766803
Pages (from-to)1911-1923
Number of pages13
JournalIEEE Journal on Selected Areas in Communications
Volume37
Issue number8
DOIs
StatePublished - Aug 1 2019

Fingerprint

Switches
Throughput
Packet loss
Network protocols
Communication
Testbeds
Simulators
Recovery

Keywords

  • queue management
  • random early detection
  • TCP incast
  • TCP timeout

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

RAPID : Avoiding TCP Incast Throughput Collapse in Public Clouds with Intelligent Packet Discarding. / Xu, Yang; Shukla, Shikhar; Guo, Zehua; Liu, Sen; Tam, Adrian S.W.; Xi, Kang; Chao, H. Jonathan.

In: IEEE Journal on Selected Areas in Communications, Vol. 37, No. 8, 8766803, 01.08.2019, p. 1911-1923.

Research output: Contribution to journalArticle

Xu, Yang ; Shukla, Shikhar ; Guo, Zehua ; Liu, Sen ; Tam, Adrian S.W. ; Xi, Kang ; Chao, H. Jonathan. / RAPID : Avoiding TCP Incast Throughput Collapse in Public Clouds with Intelligent Packet Discarding. In: IEEE Journal on Selected Areas in Communications. 2019 ; Vol. 37, No. 8. pp. 1911-1923.
@article{60da5cb6a08344bba13e52f4b4f7b8d3,
title = "RAPID: Avoiding TCP Incast Throughput Collapse in Public Clouds with Intelligent Packet Discarding",
abstract = "Many applications in public clouds require a high fan-in, many-to-one type of data communication (known as TCP incast) in modern Data Center Networks (DCNs). Such communication could cause severe incast congestion in switches and result in TCP throughput collapse, substantially degrading the application performance. The root cause of throughput collapse is the Retransmission Timeouts (RTO) due to packet losses in congested switches. Tenants in public clouds can opt to use a variety of TCP versions. However, the existing solutions rely on modifications of TCP protocols and specific techniques from switches, and thus these existing solutions are not always feasible for public clouds. In this paper, we are inspired by the emerging virtualization and network softwarization technologies to develop a novel scheme called Retransmission timeout Avoidance by Packet Intelligent Discarding (RAPID) using software switches. RAPID considers the number of packets of each incast flow, buffered in the switch to selectively discard some packets, and ensures that the Fast Retransmission/Fast Recovery rather than RTO is invoked at the sender(s) in response to packet loss. Thus, the long idle period of a timeout and the throughput drop are avoided. We prove that, given a predetermined minimum switch buffer space, dedicated to the incast application, RAPID can prevent RTO in all the incast senders. We also present a low-complexity heuristic version of RAPID named RAPID-ED, which combines the principles of RAPID and early detection and is extremely easy to implement on today's software switches. We evaluate the two proposed schemes in a data center network testbed built on NS-3 simulator. The simulation results confirm the theoretical expectation, and show that the RAPID and RAPID-ED perform very well to prevent RTO of TCP incast flows and hence the throughput collapse. Compared with other incast solutions, RAPID and RAPID-ED do not modify TCP protocols and therefore are more suitable in public clouds.",
keywords = "queue management, random early detection, TCP incast, TCP timeout",
author = "Yang Xu and Shikhar Shukla and Zehua Guo and Sen Liu and Tam, {Adrian S.W.} and Kang Xi and Chao, {H. Jonathan}",
year = "2019",
month = "8",
day = "1",
doi = "10.1109/JSAC.2019.2927072",
language = "English (US)",
volume = "37",
pages = "1911--1923",
journal = "IEEE Journal on Selected Areas in Communications",
issn = "0733-8716",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - RAPID

T2 - Avoiding TCP Incast Throughput Collapse in Public Clouds with Intelligent Packet Discarding

AU - Xu, Yang

AU - Shukla, Shikhar

AU - Guo, Zehua

AU - Liu, Sen

AU - Tam, Adrian S.W.

AU - Xi, Kang

AU - Chao, H. Jonathan

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Many applications in public clouds require a high fan-in, many-to-one type of data communication (known as TCP incast) in modern Data Center Networks (DCNs). Such communication could cause severe incast congestion in switches and result in TCP throughput collapse, substantially degrading the application performance. The root cause of throughput collapse is the Retransmission Timeouts (RTO) due to packet losses in congested switches. Tenants in public clouds can opt to use a variety of TCP versions. However, the existing solutions rely on modifications of TCP protocols and specific techniques from switches, and thus these existing solutions are not always feasible for public clouds. In this paper, we are inspired by the emerging virtualization and network softwarization technologies to develop a novel scheme called Retransmission timeout Avoidance by Packet Intelligent Discarding (RAPID) using software switches. RAPID considers the number of packets of each incast flow, buffered in the switch to selectively discard some packets, and ensures that the Fast Retransmission/Fast Recovery rather than RTO is invoked at the sender(s) in response to packet loss. Thus, the long idle period of a timeout and the throughput drop are avoided. We prove that, given a predetermined minimum switch buffer space, dedicated to the incast application, RAPID can prevent RTO in all the incast senders. We also present a low-complexity heuristic version of RAPID named RAPID-ED, which combines the principles of RAPID and early detection and is extremely easy to implement on today's software switches. We evaluate the two proposed schemes in a data center network testbed built on NS-3 simulator. The simulation results confirm the theoretical expectation, and show that the RAPID and RAPID-ED perform very well to prevent RTO of TCP incast flows and hence the throughput collapse. Compared with other incast solutions, RAPID and RAPID-ED do not modify TCP protocols and therefore are more suitable in public clouds.

AB - Many applications in public clouds require a high fan-in, many-to-one type of data communication (known as TCP incast) in modern Data Center Networks (DCNs). Such communication could cause severe incast congestion in switches and result in TCP throughput collapse, substantially degrading the application performance. The root cause of throughput collapse is the Retransmission Timeouts (RTO) due to packet losses in congested switches. Tenants in public clouds can opt to use a variety of TCP versions. However, the existing solutions rely on modifications of TCP protocols and specific techniques from switches, and thus these existing solutions are not always feasible for public clouds. In this paper, we are inspired by the emerging virtualization and network softwarization technologies to develop a novel scheme called Retransmission timeout Avoidance by Packet Intelligent Discarding (RAPID) using software switches. RAPID considers the number of packets of each incast flow, buffered in the switch to selectively discard some packets, and ensures that the Fast Retransmission/Fast Recovery rather than RTO is invoked at the sender(s) in response to packet loss. Thus, the long idle period of a timeout and the throughput drop are avoided. We prove that, given a predetermined minimum switch buffer space, dedicated to the incast application, RAPID can prevent RTO in all the incast senders. We also present a low-complexity heuristic version of RAPID named RAPID-ED, which combines the principles of RAPID and early detection and is extremely easy to implement on today's software switches. We evaluate the two proposed schemes in a data center network testbed built on NS-3 simulator. The simulation results confirm the theoretical expectation, and show that the RAPID and RAPID-ED perform very well to prevent RTO of TCP incast flows and hence the throughput collapse. Compared with other incast solutions, RAPID and RAPID-ED do not modify TCP protocols and therefore are more suitable in public clouds.

KW - queue management

KW - random early detection

KW - TCP incast

KW - TCP timeout

UR - http://www.scopus.com/inward/record.url?scp=85069926863&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069926863&partnerID=8YFLogxK

U2 - 10.1109/JSAC.2019.2927072

DO - 10.1109/JSAC.2019.2927072

M3 - Article

AN - SCOPUS:85069926863

VL - 37

SP - 1911

EP - 1923

JO - IEEE Journal on Selected Areas in Communications

JF - IEEE Journal on Selected Areas in Communications

SN - 0733-8716

IS - 8

M1 - 8766803

ER -