Huge amounts of data
The amount of data in science increases ever faster. Many research projects have hundreds of thousands of data files with information that both conclusions and further research are based on. It is evident that storing these files is not enough - they must be stored in a way that permits the information to be found if needed at a later stage - either by the project that collected or generated the data, or by others testing results or using the data for further research. At USIT we are working on new solutions for storing and refinding data.
From now on, its' FAIR
– Since I started at USIT 4 years ago, genuinely FAIR data and the solutions nedded to produce them has been a subject of conversations and dreams, says Anne Schad Bergsaker. Bergsaker is Senior Engineer at USIT and has managed the project FAIR@uio.
FAIR is an acronym for Findable, Accessible, Interoperable og Reusable. The FAIR principles for scientific data were agreed upon in 2016, and has become an internationally recognized ideal. Scientific data must be accessible, findable and reusable. "Interoperable" means that both data and metadata should be mechanically processable, and tagging and metadata should use consistent vocabularies..
USIT and UiO in the lead
UiO, as the first in Norway and one of the first in Europe and the world, now launches a platform that enables scientists to maka data genuinely FAIR. Others may store research data and call it FAIR, but without good tagging and metadata, the data is not genuinely findable, considering the vast amounts of data out there.
Project Manager Bergsaker continues: – When UiO's solution for FAIR research data is operational, it is not just for UiO researchers. We can be an important partner for both national and international projects.
UiO and USIT offers several services both within and outside the University/University College sector. The FAIR solution will be an important part of this portfolio of services available to interested parties outside UiO.
Great potential
Bergsaker tells more about the project and the platform: - Many have probably wished for a FAIRifier - a tool making their data FAIR. We're not there yet, but we are getting close. Our platform is IBM Elastic Storage Server ESS with the IBM Spectrum Discover software. Spectrum Discover uses automation and scripting to analyze the files and add correct metadata. It will also be possible for researchers to add data manually, but our goal is to have as much as possible done automatically.
- As a flagship customer of IBM we can also contribute to the further development of these solutions, so this is really exciting. We believe that the next step might be an AI running the analysis, which will make it ever more automatical and precise.
UiO's server solution today can store about 10 petabytes of data. If necessary, it can be expanded with more than 100 times that amount.
Saves research time and resources
Before further development and upscaling can take place, the platform must be operational and tested. An introductionary project is in place, and two research projects - DYNAKO at the Departement of Teacher Education and School Research and VAVU at RITMO at the Departement for Musicology.
– A solution for FAIR research data is long awaited and popular. FAIR is now a standard requirement from more and more financiers of research projects. When UiO now offers FAIR, we have become a very attractive partner, Bergsaker continues.
The usefulness is obvious and indisputable. When UiO's solution is ready, it will contain an open search - not in the research data themselves, but in the information about them. Researchers from all over the world can then find useful data for their projects and request them. Thus, Norwegian research on for instance Covid-19 can be reused in other projects. When basic data and other research data can be reused, time and resources will be saved.