1: \begin{abstract}

2: In this paper, we provide a novel matrix-analytic approach for studying doubly

3: exponential solutions of randomized load balancing models (also known as

4: \emph{supermarket models}) with Markovian arrival processes (MAPs) and

5: phase-type (PH) service times. We describe the supermarket model as a system

6: of differential vector equations by means of density dependent jump Markov

7: processes, and obtain a closed-form solution with a doubly exponential

8: structure to the fixed point of the system of differential vector equations.

9: Based on this, we show that the fixed point can be decomposed into the product

10: of two factors inflecting arrival information and service information, and

11: further find that the doubly exponential solution to the fixed point is not

12: always unique for more general supermarket models. Furthermore, we analyze the

13: exponential convergence of the current location of the supermarket model to

14: its fixed point, and apply the Kurtz Theorem to study density dependent jump

15: Markov process given in the supermarket model with MAPs and PH service times,

16: which leads to the Lipschitz condition under which the fraction measure of the

17: supermarket model weakly converges the system of differential vector

18: equations. This paper gains a new understanding of how workload probing can

19: help in load balancing jobs with non-Poisson arrivals and non-exponential

20: service times.

21: 

22: \vskip                                                      0.5cm

23: 

24: \noindent\textbf{Keywords:} Randomized load balancing, supermarket model,

25: matrix-analytic approach, doubly exponential solution, density dependent jump

26: Markov process, Markovian Arrival Process (MAP), phase-type (PH) distribution,

27: fixed point, exponential convergence, Lipschitz condition.

28: 

29: \end{abstract}