Performance and Cost Driven Data Storage and Processing for IoT Context Management Platforms

In this dissertation, we address the challenges of building a context storage management system (CSMS) as a core component of the Context as a Service (CoaaS) platform. The requirements to context query processing time, service discovery and selection, data encapsulation and overall efficiency determine the need for organizing an internal horizontally scalable and high-performing storage subsystem as a part of CoaaS platform. This research is concerned with how to architect a scalable context storage management system that can cost-efficiently and effectively respond to context queries from CoaaS platform, maintain the agreed quality of service, and provide proactive adaptation and caching.