Large intelligent network modeling using M/M/2/K/K system

With the wide deployment of intelligent network (IN) services, there is an urgent need to understand and solve teletraffic performance issues of the evolving network intelligence platform. This paper discusses a queuing system model for the performance analysis of IN call processing. The IN is presented as a network of queues where the total number of customers (e.g., SSPs) is fixed, thus forming a closed queuing network. The IN distributed architecture is modeled as a finite source queuing model - M/M/2/K/K. The expected response time for that model is analyzed and computed. The numerical results and the corresponding curves are provided. And, related to open questions, future work is summarized.