Database expert Michael Stonebraker, one of the pioneers of the relational database model, continues to bash the way relational database management systems store data.
Stonebraker is regarded as something of a database visionary after having worked in the industry for over 30 years. As a researcher at the University of California, Berkeley, in the early 1970s, he co-created the Ingres and Postgres technology that underlies many leading relational databases today, including Microsoft's SQL Server, Sybase's Adaptive Server Enterprise, Ingres's eponymous product and IBM's Informix.
But back in September last year, he broke ranks with his past, labelling relational databases legacy technology, while touting the offering from his own start-up, Vertica Systems.
More recently however, Stonebraker expanded upon his claim that there is a better way to build a data warehouse, in response to a blog by analyst Curt Monash.
Stonebraker believes that much faster response times can be achieved through a column layout (as found in Vertica's offering), instead of vertical columns and horizontal rows found in traditional relational databases.
"First, I see two categories of relational analytic/data warehouse databases, row stores and column stores," Stonebraker said in The Database Column, a website that contains blogs by database experts.
"I expect the overwhelming majority of analytic data management workloads to move to column stores over time as these products become more mature because of the overwhelming performance advantage they offer on most analytic workloads," he wrote.
Stonebraker also said that if performance was not a big issue, then current open-source relational database management systems would work quite well. "As a result, I expect the 'low end' to go to open source systems."