Collaborative Research: SaTC 2.0: RES: Automating Access Control Enforcement for Decentralized Microservice Systems

Cloud computing systems have evolved from monolithic servers to distributed, service-oriented systems, called microservice systems, that enable scalable, flexible, and rapidly evolving applications through independently deployable services. While microservice systems offer potential benefits in scalability, resilience, and ease of deployment, they have also introduced vulnerabilities in access control enforcement that directly undermine the isolation and least-privilege assumptions central to the microservice paradigm. These access control enforcement vulnerabilities emerge because: (i) the parties that define the access control policies differ from the parties that enforce the policies, and (ii) current access control enforcement does not account for runtime changes that may result in unauthorized access. To make matters worse, with the rise of agentic microservices, access control becomes yet more complex due to the semantic gap between the user’s intent and the agent's actual execution. This project develops theories and techniques for deploying next-generation microservices in a manner that enforces access control effectively and efficiently to protect microservice data security. The project's novelties are in its development of principled approaches for systematic vulnerability testing of microservice systems and in the automated approaches for hardening access control enforcement based on the test results. The project's broader significance and importance are the improvement of the security and resilience of a multitude of microservice applications, such as for healthcare, finance, and transportation, as well as the release of open-source tools to benefit the broader community and the training of the next generation of cybersecurity researchers and practitioners. This project develops theories and techniques to improve access control enforcement both within microservices (i.e., the service layer) and for their deployment on hosts (i.e., the system layer) to protect microservice data security. First, this project develops a principled testing infrastructure that identifies the properties to test to detect vulnerabilities in access control enforcement for traditional and agentic microservice systems. Second, this project develops novel mechanisms to test the system-layer operations that may impact access control enforcement, including microservice orchestration and container deployment. Third, this project investigates how to apply these generated test results (i.e., the discovery of vulnerabilities) automatically to harden the service and system layers through the software retrofitting of access control enforcement mechanisms and the addition of the necessary context to the access control policies to block vulnerabilities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.