The Busy Period analysis of Queueing Systems sheds light on the transient nature of the system. Due to the complexities that are involved in analyzing transient systems, not many results for the Busy Periods are known outside the domain of Markovian systems. In this talk we first briefly present a couple of results related to the number of customers served during a busy period and the lengths of those busy periods in a G/G/1 system, where both the arrival and the service processes can be serially correlated. A recursive solution to the exact probabilities of 'n' number of customers being served in a G/G/1 busy period is provided and derivations for its moments are given. The time distributions for the possible paths during a Busy Period is also characterized as an Matrix Exponential process. We then present the effect of increase in threshold level and the correlations in the arrival and service processes on the mean first passage time to go below the given threshold level. Finally we look at the probabilities that a max height of 'k' is reached during a busy period.
Chaitanya Garikiparthi completed his M.S in Computer Science from the Univ. of Texas at Dallas in 2001 and is currently working on his Ph.D. at the Univ. of Missouri - Kansas City. He did a couple of internships in the industry (Nortel Networks in the Summer of 2001 and at Motorola in Summer of 2006). His main research interests are Queueing Theory, Modelling and Performance Analysis of Computer systems and networks.