Performance Measures of Priority Queuing Data Network Node with Unreliable Transmission Channel

Abstract

We propose analysis of the priority queuing network node, which is using for transmission unreliable channel with constant bit transmission rate. For queuing system with two separated finite capacity buffers for each priority data packet flows we investigate performance measures in two cases: when transmission channel is reliable and when transmission channel is unreliable. The inter arrival time of high and low priority data packet flows and the transmission times of each priority data packet are exponentially distributed. The arrival process of each priority data packet is assumed to be independent of each other and the transmission process. Data packet leaves the buffer in first come first service manner. Non-preemptive priority queuing model is analyzed. Research is in progress to develop analytical and simulation models for such network data packets transmission node. Using our proposed analytical models we can easily evaluate system performance measures for each priority data packet, such as data packet losses, mean value of waiting time in buffer, mean number of data packets in queue, mean number of data packets in the system. The results from analytical and simulation model of some performance measures are taken in figures. The processes in the priority queuing network node are based on the Markov chains. The proposed analytical system models are suitable for analysis network node with finite data packets buffer. When the system buffer capacity is high analytical model for such queuing system node becomes very complicated. Simulation method allows investigate system with large capacities and exponential or constant data packets arrival process when transmission time so is exponential or constant. Our proposed analytical and simulation models would be very worthwhile to analyze the processes on real-time services providing data network node. Ill. 7, bibl. 5 (in English; summaries in English, Russian and Lithuanian).