SIParCS 2020 - Ela Jamali
Object-Oriented Software Development with Agile Scrum
Easily accessible climate data is a valuable resource that allows climate scientists, graduate students, and citizen scientists to perform their own science and move science forward as a whole. Historically, NCAR has not had great processes or tools for exposing NCAR created digital scientific assets like models, simulations, historical data, papers, and software to the public. To improve public access and fulfill requirements from the NSF, it was decided to create a web-based search tool that allows end-users to search across all digital scientific assets available at NCAR. The Metadata Search application, a multi-service application, was developed with two services built on top of Solr. The Harvester service reads all the XML files provided by different labs at NCAR and inserts various fields into Solr. Additionally, the Search service queries from Solr and provides a simplistic User Interface to end-users. Both Services are independently deployable and are orchestrated and maintained with Docker Swarm. Through the usage of agile scrum methodologies, a minimal viable product was created for this SIParCS project in 10 weeks.
Mentors: Nathan Hook, Christy Grant, Eric Nienhouse
Slides