Fabricio Bandeira Cabral
Published 2007-06-04Version 1
In this paper, we consider systems that can be modelled by $M \mid M \mid n$ queues with heterogeneous servers and non informed customers. Considering any two servers: we show that the probability that the fastest server is busy is smaller than the probability that the slowest server is busy. Moreover, we show that the effective rate of service done by the fastest server is larger than effective rate of service done by the slowest server.
Related articles:
Explicit Steady-State Approximations for Parallel Server Systems with Heterogeneous Servers
Approximate Local Limit Theorems with Effective Rate and Application to Random Walks in Random Scenery
Many Server Queueing Models with Heterogeneous Servers and Parameter Uncertainty with Customer Contact Centre Applications
![QR code for arXiv:0706.0560 [math.PR]](http://oss.arxitics.com/images/favicon.svg)