The Centralized connection management method is vulnerable due to its single point of failure and also as the traffic load increases, the control traffic to and from the controller increases substantially and the central controller requires sufficient buffer and processing speed to handle the requests. To overcome the above limitation, we present the design and analysis of a Priority based efficient and reliable centralized connection management scheme with two central management systems (CMS-1 and CMS-2) for computer networks (PERCC). In this scheme whenever a request comes, the odd-numbered requests will be taken up by CMS-1 and also stored in a queue of CMS-2 and vice-versa for the even-numbered requests. CMS-1 and CMS-2 check for any high priority request among three requests in the incoming buffer queue, when taking up. If the request-taken CMS cannot allocate a path, then the other CMS will automatically take it up after t seconds in case if it's a low priority request, or before t1 seconds if it's a high priority request (t1 < t), which gives a second chance for any failed request for getting a path. The timer for high priority requests is less than low priority ones. They will be attended faster/before than low priority requests. Furthermore, if any one of the CMS fails, then the other CMS will take over the entire work. By simulation results, we showed that our proposed centralized connection management scheme has low blocking probability and processing time in comparison with the single CMS connection management scheme. © 2011 IEEE.
cited By (since 1996)0; Conference of org.apache.xalan.xsltc.dom.DOMAdapter@3ecf3c10 ; Conference Date: org.apache.xalan.xsltc.dom.DOMAdapter@17705ea7 Through org.apache.xalan.xsltc.dom.DOMAdapter@525631f6; Conference Code:89416
S. Kashyap, Suman, I. V. S. A., Ramesh, T. K., and Vaya, P. R., “Priority based efficient and reliable centralized connection management scheme for computer networks”, in TISC 2011 - Proceedings of the 3rd International Conference on Trendz in Information Sciences and Computing, Chennai, 2011, pp. 169-172.