TY - JOUR

T1 - System crash in a finite capacity M/G/1 queue

AU - Knessl, C.

AU - Matkowsky, B. J.

PY - 1986

Y1 - 1986

N2 - We analyze a finite capacity M/G/1 queue in which an arrival that causes the system capacity of unfinished work to be exceeded, results in the loss of all jobs in the system. We sbudy the stationary density of unfinished work, the system utilization, the mean length of the busy period, and the probability that the system crashes before completing a busy period. We obtain formulas for these quantities for an M/M/1 queue with constant arrival and service rates. Approximate expressions for these quantities are constructed for M/G/1 queues in which the arrival rates and service densities depend on the amount of unfinished work in the system at the instant a new customer enters. These approximations, obtained by using singular perturbation techniques, are shown to agree with the exact results for state-dependent M/M/1 queues.

AB - We analyze a finite capacity M/G/1 queue in which an arrival that causes the system capacity of unfinished work to be exceeded, results in the loss of all jobs in the system. We sbudy the stationary density of unfinished work, the system utilization, the mean length of the busy period, and the probability that the system crashes before completing a busy period. We obtain formulas for these quantities for an M/M/1 queue with constant arrival and service rates. Approximate expressions for these quantities are constructed for M/G/1 queues in which the arrival rates and service densities depend on the amount of unfinished work in the system at the instant a new customer enters. These approximations, obtained by using singular perturbation techniques, are shown to agree with the exact results for state-dependent M/M/1 queues.

UR - http://www.scopus.com/inward/record.url?scp=84935739891&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84935739891&partnerID=8YFLogxK

U2 - 10.1080/15326348608807032

DO - 10.1080/15326348608807032

M3 - Article

AN - SCOPUS:84935739891

SN - 0882-0287

VL - 2

SP - 171

EP - 201

JO - Communications in Statistics. Stochastic Models

JF - Communications in Statistics. Stochastic Models

IS - 2

ER -