New array database technology for scientists

Alex van Ballegooij received his PhD degree for his research on a new array database technology for scientists on 17 September.

Scientific research and simulations produce enormous amounts of data that need to be organized, managed and analysed. Although scientists could benefit from database technology, they hardly use it. One of the reasons is that scientific data are stored in other structures than the structure database systems offer. Alex van Ballegooij, PhD at the Centrum Wiskunde & Informatica (CWI) in Amsterdam, developed a new language, RAM, that connects research data and relational databases. Van Ballegooij received his PhD on Wednesday, 17 September at the University of Amsterdam on his thesis: 'RAM: Array Database Management through Relational Mapping'. Objective of his research is to realize an array database-architecture. Internationally this topic receives a lot of attention.

In contrast to former initiatives to develop an array database architecture, the approach of Van Ballegooij does not require designing a whole new database system, nor to adapt an existing system. The researcher reuses existing technologies by means of relational mapping: representing new structures and operations in a language that can be understood by a relational database management system. He uses array-structures: rectangular structures, like a matrix, cube or a block. Scientific data and multimedia are easier to store and to query in these structures than in traditional records. The computer scientist implemented his method on the MonetDB-database system that is also developed by CWI. It was tested using a multimedia search application.

Until now, scientists often developed their own systems for data storage. The new system allows scientists to benefit from innovations and benefits of database technology, just as the financial world already does. “This kind of research might help database technology to get influence in the scientific world”, says Van Ballegooij. Controlling the data explosion is an important theme of CWI. This research is a good example.

More information:
- the thesis; supervisor: Prof. dr. M.L. Kersten (CWI and UvA), co-supervisor: Prof. dr. ir. A.P. de Vries (CWI and TU Delft); PhD defence: 17 September 2009 at 14.00h, Agnietenkapel, Oudezijds Voorburgwal 231, Amsterdam
- The research is co-financed by the MIA and MultimediaN projects
- Picture: Shutterstock