DocumentCode :
265873
Title :
Admission control with flow aggregation for QoS provisioning in software-defined network
Author :
Jun Huang ; Yunlong He ; Qiang Duan ; Qing Yang ; Wei Wang
Author_Institution :
Sch. of Commun. & Inf. Eng., Chongqing Univ. of Posts & Telecommun., Chongqing, China
fYear :
2014
fDate :
8-12 Dec. 2014
Firstpage :
1182
Lastpage :
1186
Abstract :
Software Defined Network (SDN) may significantly enhance network and service management by enabling separated control and data planes. The centralized OpenFlow controller with a global vision of network states offers a promising approach to realizing flow-based admission control for supporting Quality of Service (QoS) provisioning in SDN. However, per-flow process brings in challenges to scalability of OpenFlow-based SDN. Flow aggregation has been explored as an effective method to address this issue. In this paper, we investigate admission control with flow aggregation for QoS provisioning in SDN. Specifically we propose a model for admission control with flow aggregation and develop the analysis techniques for determining the required amounts of bandwidth and buffer space at OpenFlow-enabled switches for meeting performance requirements in delay and packet loss. Network calculus is applied in our modeling and analysis; which makes our method applicable to general OpenFlow-based SDNs with various implementations. Numerical experiment results are also provided to evaluate effectiveness of the developed modeling and analysis techniques.
Keywords :
protocols; quality of service; queueing theory; software defined networking; OpenFlow controller; OpenFlow-enabled switch; QoS provisioning; SDN; admission control; buffer space; flow aggregation; network calculus; packet loss; per-flow process; quality of service; queuing systems theory; service management; software-defined network; Admission control; Aggregates; Analytical models; Bandwidth; Control systems; Delays;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Global Communications Conference (GLOBECOM), 2014 IEEE
Conference_Location :
Austin, TX
Type :
conf
DOI :
10.1109/GLOCOM.2014.7036969
Filename :
7036969
Link To Document :
بازگشت