Database Preservation Toolkit: a flexible tool to normalize and give access to databases

Abstract

Digital preservation is emerging as an area of work and research that tries to provide answers that will ensure a continued and long-term access to information stored digitally. IT Platforms are constantly changing and evolving and nothing can guarantee the continuity of access to digital artifacts in their absence. This paper focuses on a specic family of digital objects: Relational Databases; they are the most frequent type of databases used by organizations worldwide. Database Preservation Toolkit enables the preservation of relational databases holding the structure and content of the the database in a preservation format in order to provide access to the database information in a long term period. If in one hand there is a need to migrate databases to newer ones that appear with technological evolution, on the other hand there is also the need to preserve the information they hold for a long time period, due to legal duties but also due to archival issues. That being said, that information must be available no matter the database management system where the information came from. In this area, solutions are still scarce. Main products for relational database preservation include CHRONOS and SIARD. The rst one is, in most of the cases, unreachable due to the associated costs. The second one is not really a product but a preservation format. The main idea behind this work was to explore the main features and limitations of the existing products in order to improve 'db-preservation-toolkit' (http://keeps.github.io/ db-preservation-toolkit/), an extracted component from the RODA project (http://www.roda-community.org). Therefore, 'db-preservation-toolkit' was improved with respect to performance and also with new features addiction in order to support more database management systems, address some missing features of the other products, support of a new preservation format (SIARD) and provide an interface where it is possible to access and search the information of the archived databases.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.