MACE: A Mobile Ad-hoc Computing Emulation Framework

Abstract

Designing and deploying mobile applications for Edge/Fog environments is challenging. Previous studies have introduced simulation and emulation tools that can help to model new applications but fail to deliver an environment in which a system designer can emulate a realistic deployment of such applications. We propose the Mobile Ad-hoc Computing Framework (MACE) 4 that provides features which facilitate and accelerate the development and testing of Edge/Fog applications. It also provides the nodes with mobility control that can be modelled or triggered by external applications, such as a flight simulator. With the included GUI, the user can plan deployment topologies, observe and interact with the nodes in real-time. To demonstrate its functionalities, two experimental evaluations were performed: one comparing the network emulators and another with an off-the-shelf configuration service running in mobile nodes. Finally, the functionality of controlling mobility via an external application is also demonstrated.

Assessing Distributed Consensus Performance on Mobile Cyber-Physical System Swarms

“…We are interested in understanding the performance of consensus protocols in mobile networks with properties similar to MANETs. In order to reproduce additional features specific to CPS systems, this work relies on MACE [15], an emulation framework for mobile CPS.…”

“…To reproduce such a mobility within A, our experiments rely on the Mobile Ad-hoc Computing Emulator [15], MACE, an emulation framework for mobile CPS swarms. MACE generates a virtual environment as close to reality as possible that allows us to run and evaluate distributed consensus implementations.…”

“…Specifically, we aim at evaluating the performance of a realworld implementation of multi-Paxos [42], by far the most popular consensus protocol in production systems. Using an emulation framework for mobile CPS [15], we investigate the impact of three practical, common uncertainties created by the mobile environment of CPS swarms: (1) the effective cost of mobility of the nodes within the deployed mission's area, (2) the impact of the connectivity of nodes and the eventual network partitions, and (3) cost of continuous routing updates in the mobile environment.…”