Simulation Environment of Evaluation and Optimization for Moving Target Defense

Abstract

In this paper, a simulation environment for evaluating the effectiveness and performance cost of moving target defense (MTD) is presented. This simulation environment is based on SimPy, which is a process-based discrete-event simulation framework worked on standard Python. In our MTD simulation environment, we develop a typical topology to contain different deployments of moving target (MT) techniques. Meanwhile, processes of attackers, defenders, services, and users are analyzed and modeled in the simulation environment, which also provides an extendable interface for MT technologies in the future. In the case study, we quantify the effectiveness and performance of MTD in the proposed simulation environment under several typical scenarios, none-MTD deployment, high-security strategy, high-performance strategy. The average attacking time, average processed requests, and average request's processing time are used as metrics to assess different MTD strategies. Results show that our simulation environment provides an efficient and intuitionistic method for evaluating the effectiveness and performance of MTD and helping to choose a more optimal strategy set for MTD.