Building accurate workload models using Markovian arrival processes

  title={Building accurate workload models using Markovian arrival processes},
  author={Giuliano Casale},
The application of Markov chains and queueing theory to real performance evaluation problems often raises the question on how to best integrate in the model the observed characteristics of a workload. For example, what if job inter-arrival times to the system are statistically correlated? How can this be described compactly in a Markov model? What if the service time distribution is heavytailed? What if arrivals are periodic? Markovian arrival processes (MAPs) offer an elegant solution to these… CONTINUE READING


Publications citing this paper.
Showing 1-4 of 4 extracted citations


Publications referenced by this paper.

Structured Stochastic Matrices of M/G/1 Type and Their Applications

M. F. Neuts
View 4 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…