Secure Platform for Processing Sensitive Data on Shared HPC Systems
This addresses the growing need for secure data processing on shared HPC systems, particularly for researchers handling sensitive information, though it appears incremental as it builds on existing virtualization technologies without modifying HPC infrastructure.
The authors tackled the problem of securely processing sensitive data on shared high-performance computing (HPC) systems by developing a platform-as-a-service solution using PCOCC and SLURM, which is scalable with low performance overhead and has been applied in real-world research applications.
High performance computing clusters operating in shared and batch mode pose challenges for processing sensitive data. In the meantime, the need for secure processing of sensitive data on HPC system is growing. In this work we present a novel method for creating secure computing environments on traditional multi-tenant high-performance computing clusters. Our platform as a service provides a customizable, virtualized solution using PCOCC and SLURM to meet strict security requirements without modifying the exist-ing HPC infrastructure. We show how this platform has been used in real-world research applications from different research domains. The solution is scalable by design with low performance overhead and can be generalized for processing sensitive data on shared HPC systems imposing high security criteria