NIRD2020 national storage infrastructure for research data

The existing national Data Storage infrastructure must be replaced by a new one, hereafter NIRD2020. NIRD2020 will serve the needs of researchers from different sectors and disciplines, therefore it will host a large variety of scientific data, differencing in file size, file count, format, retention policy, integrity requirements, access frequency, etc. NIRD2020 needs to accommodate not only traditional compute-intensive users, but also data-driven user communities. Data stored in NIRD2020 might come from different sources: might be produced by projects through analysis on hpc systems, or directly ingested into the storage system from labs or remote sensors, or might be sourced from local and cloud repositories. The large amount of data produced will be ingested via effective protocols and stored safely to assure data integrity. NIRD2020 will be reliable, flexible, and scalable in all components, both in terms of capacity and capability. Also, NIRD2020 will support multitenancy in order to accommodate the relevant users and research communities.