Liquid Computing on Multiclustered Hybrid Environment for Data Protection and Compliance

In today's world, the spread of personal data has become a major concern for individuals and organizations. With the increasing amount of data being gathered and stored, it is increasingly essential to regulate the processes of data collection and processing. The EU's General Data Protection Regulation (GDPR) is an example of such regulation, which mandates the protection of personal data of EU citizens. This thesis aims to analyse and propose a solution to the problem of using personal data within distributed systems called ecosystem platforms. For ecosystem platforms is intended a type of digital platform that provides a set of core technologies and services upon which other businesses can build their own products, services, or platforms. Ecosystem platforms typically offer a suite of APIs, software development kits (SDKs), and other tools that allow third-party developers to access the platform's functionality and create their own applications or services that can integrate with the platform. The cloud-native approach is particularly leveraged by these platforms, which make extensive use of containers and orchestrators. Containerization is a method of packaging software that enables applications to run reliably and consistently across different computing environments and this is achieved by encapsulating the application code, libraries, and dependencies into a single package, known as a container. An orchestrator is a tool that automates the deployment, scaling, and management of containerized applications across a cluster. By using containers and orchestrators in a cloud-native approach, the proposed solution can provide a scalable, portable, and flexible architecture that can adapt to the evolving needs of the ecosystem platform. This thesis focuses on the problem of using ecosystem platforms in distributed scenarios around the world, which must comply with different regulations depending on the customer's location. In particular, the aim is to propose a solution that does not alter the platforms and allows the continued use of containerized approaches, while simplifying the management of the data that should be customized for each location to comply with regulations. The proposed solution is to use liquid computing, a concept that involves creating a continuum of computing resources that can be used to support various types of workloads. This approach involves combining centralized cloud resources with distributed resources that are located closer to the edge of the network, such as on devices or in local data centres with the overall objective of enabling organizations to leverage the full potential of the cloud while maintaining control over their data and ensuring compliance with data privacy regulations. This approach is optimal for the highlighted problem as it allows for the creation of a single point of control for a hybrid infrastructure, providing the ability to perform storage and processing of data locally. The entire solution is offered transparently, meaning that both the end user and the developer can utilize the devices at the edge for data storage without being aware of it. The thesis proposes a specific use case that deals with creating databases that can be allocated both in the central cluster and in the various connected edge devices. This proof-of-concept makes the validity of this solution evident and also demonstrates its applicability through tests that have been carried out in order to make it part of the European lighthouse project Gaia-X.