Analysis on Buffer Occupancy of Quantized Congestion Notification in Data Center Networks

Chang RUAN; Jianxin WANG; Jiawei HUANG; Wanchun JIANG

doi:10.1587/transcom.2016EBP3052

Analysis on Buffer Occupancy of Quantized Congestion Notification in Data Center Networks

Chang RUAN, Jianxin WANG, Jiawei HUANG, Wanchun JIANG

Full Text Views

0

Share
Cite this

Summary :

In data center networks, Quantized Congestion Notification (QCN) has been ratified as the standard congestion management mechanism in the link layer. Since QCN nonlinearly switches between the rate increase and decrease stages, it is very difficult to understand QCN in depth and provide theoretical guidelines on setting the buffer size of the QCN switch. This paper gives an explicit formula for the boundary of buffer occupancy of the QCN switch. Specifically, based on the fluid model of QCN, we first derive the uniformly asymptotic stability condition of the QCN system. Then, under the condition that QCN is uniformly asymptotically stable, we analyze the buffer occupancy of the QCN switch theoretically and show that the classic rule-of-thumb for buffer sizing is not suitable for QCN. Finally, simulations validate the accuracy of our theoretical results.

Publication: IEICE TRANSACTIONS on Communications Vol.E99-B No.11 pp.2361-2372

Publication Date: 2016/11/01

Publicized: 2016/06/01

Online ISSN: 1745-1345

DOI: 10.1587/transcom.2016EBP3052

Type of Manuscript: PAPER

Category: Network

Authors

Chang RUAN
  Central South University
Jianxin WANG
  Central South University
Jiawei HUANG
  Central South University
Wanchun JIANG
  Central South University

Keyword

quantized congestion notification, buffer occupancy, data center networks, nonlinear

Cite this

Copy

Chang RUAN, Jianxin WANG, Jiawei HUANG, Wanchun JIANG, "Analysis on Buffer Occupancy of Quantized Congestion Notification in Data Center Networks" in IEICE TRANSACTIONS on Communications, vol. E99-B, no. 11, pp. 2361-2372, November 2016, doi: 10.1587/transcom.2016EBP3052.
Abstract: In data center networks, Quantized Congestion Notification (QCN) has been ratified as the standard congestion management mechanism in the link layer. Since QCN nonlinearly switches between the rate increase and decrease stages, it is very difficult to understand QCN in depth and provide theoretical guidelines on setting the buffer size of the QCN switch. This paper gives an explicit formula for the boundary of buffer occupancy of the QCN switch. Specifically, based on the fluid model of QCN, we first derive the uniformly asymptotic stability condition of the QCN system. Then, under the condition that QCN is uniformly asymptotically stable, we analyze the buffer occupancy of the QCN switch theoretically and show that the classic rule-of-thumb for buffer sizing is not suitable for QCN. Finally, simulations validate the accuracy of our theoretical results.
URL: https://globals.ieice.org/en_transactions/communications/10.1587/transcom.2016EBP3052/_p

Copy

@ARTICLE{e99-b_11_2361,
author={Chang RUAN, Jianxin WANG, Jiawei HUANG, Wanchun JIANG, },
journal={IEICE TRANSACTIONS on Communications},
title={Analysis on Buffer Occupancy of Quantized Congestion Notification in Data Center Networks},
year={2016},
volume={E99-B},
number={11},
pages={2361-2372},
abstract={In data center networks, Quantized Congestion Notification (QCN) has been ratified as the standard congestion management mechanism in the link layer. Since QCN nonlinearly switches between the rate increase and decrease stages, it is very difficult to understand QCN in depth and provide theoretical guidelines on setting the buffer size of the QCN switch. This paper gives an explicit formula for the boundary of buffer occupancy of the QCN switch. Specifically, based on the fluid model of QCN, we first derive the uniformly asymptotic stability condition of the QCN system. Then, under the condition that QCN is uniformly asymptotically stable, we analyze the buffer occupancy of the QCN switch theoretically and show that the classic rule-of-thumb for buffer sizing is not suitable for QCN. Finally, simulations validate the accuracy of our theoretical results.},
keywords={},
doi={10.1587/transcom.2016EBP3052},
ISSN={1745-1345},
month={November},}

Copy

TY - JOUR
TI - Analysis on Buffer Occupancy of Quantized Congestion Notification in Data Center Networks
T2 - IEICE TRANSACTIONS on Communications
SP - 2361
EP - 2372
AU - Chang RUAN
AU - Jianxin WANG
AU - Jiawei HUANG
AU - Wanchun JIANG
PY - 2016
DO - 10.1587/transcom.2016EBP3052
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E99-B
IS - 11
JA - IEICE TRANSACTIONS on Communications
Y1 - November 2016
AB - In data center networks, Quantized Congestion Notification (QCN) has been ratified as the standard congestion management mechanism in the link layer. Since QCN nonlinearly switches between the rate increase and decrease stages, it is very difficult to understand QCN in depth and provide theoretical guidelines on setting the buffer size of the QCN switch. This paper gives an explicit formula for the boundary of buffer occupancy of the QCN switch. Specifically, based on the fluid model of QCN, we first derive the uniformly asymptotic stability condition of the QCN system. Then, under the condition that QCN is uniformly asymptotically stable, we analyze the buffer occupancy of the QCN switch theoretically and show that the classic rule-of-thumb for buffer sizing is not suitable for QCN. Finally, simulations validate the accuracy of our theoretical results.
ER -