Identifier Services: Modeling and Implementing Distributed Data Management in Cyberinfrastructure

Abstract

The Identifier Services (IDS) project conducted research into and built a prototype to manage distributed genomics datasets remotely and over time. Inspired by archival concepts, IDS allows researchers to track dataset evolution through multiple copies, modifications, and derivatives, independent of where data are located – both symbolically, in the research lifecycle, and physically, in a repository or storage facility. The prototype implementation is based on a three-step data modeling process involving: a) understanding and recording of different researcher workflows, b) mapping the workflows and data to a generic data model and identifying functions, and c) integrating the data model as architecture and interactive functions into cyberinfrastructure (CI). Identity functions are operationalized as continuous tracking of authenticity attributes including data location, differences between seemingly identical datasets, metadata, data integrity, and the roles of different types of local and global identifiers used during the research lifecycle. CI resources were used to conduct identity functions at scale, including scheduling content comparison tasks on high-performance computing resources. The prototype was developed and evaluated considering six data test cases, and feedback was received through a focus-group activity. While there are some technical roadblocks to overcome, our project demonstrates that identity functions are innovative solutions to manage large distributed genomic datasets.

A Flexible, Extensible, Machine-Readable, Human-Intelligible, and Ontology-Agnostic Metadata Schema (OIMS)

“…A Data Entity, is a top level container of information (Esteva et al, 2019). From a machine perspective it is most relevant as a data object that has a unique uniform resource identifier (URI) (Berners-Lee et al, 2005) and at least some technical metadata.…”

Curation and Publication of Simulation Data in DesignSafe, a Natural Hazards Engineering Open Platform and Repository

“…In DS, we saw the need and the opportunity to develop a model specifically to represent all the components and complex structure of simulation data. The methods used to develop it were similar to those followed to design experimental and field research models and services in DS [13] and in other data cyberinfrastructure projects [11,14].…”