Ceph at the Digital Repository Ireland (DRI)

Presenters:
Peter Tiernan - TCD

Description:
This presentation addresses the themes of real world scalability & performance, open technologies, digital preservation tools, services & infrastructure by discussing open storage solutions.

The Digital Repository Of Ireland (DRI) is an interactive, national trusted digital repository for contemporary and historical, social and cultural data held by Irish institutions. The DRI has a requirement for a massively scalable, fully featured storage solution to store and preserve digital assets. Software defined storage has brought scalable storage within the reach of smaller projects like the DRI . The power and features provided by such systems can prove revolutionary and has the potential to deliver stable, vastly scalable storage to repository infrastructures at low cost.

This presentation explores four such solutions: Ceph, iRODS, GPFS and HDFS. Their features were analysed and tested to assess suitability for the DRI. Criteria useful to digital repositories such as high availability, interfaces, interoperability, replication and data security were used to determine which solution best fits. The chosen solution was integrated into the DRI infrastructure. The DRI use case is described along with comments on implementation and performance.