A Provenance Management System for Research Data Management in High-Performance Computing Systems

Research data management (RDM) has become increasingly important recently. High-performance computing (HPC) systems that process large volumes of data face the problem of how to manage the research data produced in their systems. Since the purposes of RDM are to improve the reproducibility of research and the reusability of the research data, collecting and managing the provenance of data produced in HPC systems is essential. In this paper, we introduce the concept of a provenance management system for RDM in HPC systems. We analyze use cases of provenance and the requirements for collecting and managing provenance. We also discuss implementation of the requirements. Finally, we introduce our prototype of the provenance management system.