| Over the past decade, as industry in nearly every sector of the economy has moved to a data-driven world, there has been an explosion in the volume of data and the need for richer and more flexible data processing tools and environments. Processing and analyzing large volumes of data has become easier over the last few years, and today there are many distributed data processing tools and cloud computing platforms that can be readily used to analyze large databases. In other words, users now have access to a variety of open-source tools for "big data" management, tailored for different types of workloads. Most of these systems can be easily deployed to large clusters and can scale to thousands of nodes. Moreover, users do not need to purchase expensive hardware to deploy their applications. There are a variety of cloud computing platforms, adjusted for different types of workloads and hardware requirements, which further simplify the process of deploying, managing and controlling a large-scale application. At the same time, the migration to a cloud-based environment can potentially reduce the end-to-end cost for the user. As a result, users are increasingly leaning towards cloud environments for data-centric applications.;Given the increasing number of available data management solutions and the current computing trend towards cloud-based systems, the work in this dissertation focuses on studying the design of different types of data processing systems that can be deployed in cloud-based environments, identifying its effects on workload performance, and proposing new ways to improve it. |
