Title :
Connection management optimization method based on network dataflow locality
Author_Institution :
Coll. of Comput. Inf. & Eng., Xinxiang Univ., Xinxiang, China
Abstract :
Targeting the performance requirements of connection management in high-speed networks, an optimization of connection management is proposed, applying Move to Front (MTF) heuristic to improve the lookup operation of connection table. The locality property of network traffic is analyzed, which is represented as a group of packets within a connection probably arrive in group in a short time. Based on the property, the MTF heuristic is applied to optimize the lookup operation of connection table, and get the MTF connection table. The implementation of optimized connection management is described. The optimization of MTF connection table is evaluated via physical high-speed network traffic. Experimental results indicate that the lookup performance of MTF connection table is superior to that of the traditional sorted connection table.
Keywords :
data communication; telecommunication network management; telecommunication traffic; MTF connection table lookup performance; MTF heuristic; connection management optimization method; high-speed networks; move to front heuristic; network dataflow locality; physical high-speed network traffic; sorted connection table; Local area networks; Sorting; Move to Front (MTF) heuristic; connection management; high-speed network; locality of network dataflow;
Conference_Titel :
Electrical & Electronics Engineering (EEESYM), 2012 IEEE Symposium on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4673-2363-5
DOI :
10.1109/EEESym.2012.6258744