DocumentCode
3214356
Title
Performance Analysis of an Efficient Flow Control Mechanism with Job Threshold in InfiniBand Networks
Author
Yan, Shihang ; Awan, Irfan ; Min, Geyong
Author_Institution
Univ. of Bradford, Bradford
fYear
2008
fDate
25-28 March 2008
Firstpage
249
Lastpage
256
Abstract
InfiniBand Architecture (IBA) is defined as a switch-based interconnection technology with high- bandwidth and low-latency. It is suitable for constructing high-speed networks for cluster systems and has been ratified as a new industry standard for the server I/O and inter-processor communication. In this paper, we propose an efficient flow control mechanism with job threshold for InfiniBand networks. This new scheme can reduce the delay, blocking probability, and mean queue length as well as enhance the Quality of Service (QoS) of InfiniBand networks. The principle of maximum entropy (ME) is adopted as an effective methodology to analyze the new mechanism with the generalized exponential (GE)-type distribution for modeling the inter-arrival times and service times of the input traffic. Moreover, typical numerical experiments are included to validate the results of ME algorithm against simulation.
Keywords
computer networks; microprocessor chips; quality of service; telecommunication congestion control; telecommunication network topology; InfiniBand architecture; InfiniBand networks; ME algorithm; blocking probability; cluster systems; flow control mechanism; generalized exponential-type distribution; high-speed networks; infiniband networks; inter-arrival times; inter-processor communication; job threshold; maximum entropy; mean queue length; quality of service; server I/O communication; service times; switch-based interconnection technology; Bandwidth; Communication industry; Communication standards; Communication switching; Communication system control; Electrical equipment industry; High-speed networks; Network servers; Performance analysis; Quality of service;
fLanguage
English
Publisher
ieee
Conference_Titel
Advanced Information Networking and Applications, 2008. AINA 2008. 22nd International Conference on
Conference_Location
Okinawa
ISSN
1550-445X
Print_ISBN
978-0-7695-3095-6
Type
conf
DOI
10.1109/AINA.2008.138
Filename
4482715
Link To Document