Persistent Identifier Practice for Big Data Management at NCI

Abstract

The National Computational Infrastructure (NCI) manages over 10 PB research data, which is co-located with the high performance computer (Raijin) and an HPC class 3000 core OpenStack cloud system (Tenjin). In support of this integrated High Performance Computing/High Performance Data (HPC/HPD) infrastructure, NCI’s data management practices includes building catalogues, DOI minting, data curation, data publishing, and data delivery through a variety of data services. The metadata catalogues, DOIs, THREDDS, and Vocabularies, all use different Uniform Resource Locator (URL) styles. A Persistent IDentifier (PID) service provides an important utility to manage URLs in a consistent, controlled and monitored manner to support the robustness of our national ‘Big Data’ infrastructure. In this paper we demonstrate NCI’s approach of utilising the NCI’s 'PID Service 'to consistently manage its persistent identifiers with various applications.

Keywords

• Persistent Identifier
• Big Data
• URI
• Data catalogue
• Data Management