A Multi-agent Persistency Framework using Cloud-Based Agent Mirroring

Abstract

This thesis introduces a exible cloud-based persistency framework that can be used in a wide range of Multi-agent systems. The recent emergence of the Internet of Things has made it possible for us to monitor and manage our environment like never before. Small embedded devices can run an array of sensors and communicate with each other using mechanisms such as Blue- tooth SMART. Alongside such hardware developments, intelligent data pro- cessing and control software also needs to be created. Multi-agent systems provide a well-tailored architecture for such challenges. They are however often limited in terms of persistency and scalability. In our work we have designed and implemented a new in memory agent persistency framework. In order to evaluate proposed approach two prototypes have been implemented, one using the proposed solution and the other using an established agents persistency environment. Experimental results con rm that our framework is more scalable than existing approaches whilst providing a similar level of persistency.