Martin L. Kersten

Martin L. Kersten (October 25, 1953 – July 6, 2022) was a computer scientist with research focus on database architectures, query optimization and their use in scientific databases. He was an architect of the MonetDB system, an open-source column store for data warehouses, online analytical processing (OLAP) and geographic information systems (GIS). He has been (co-) founder of several successful spin-offs of the Centrum Wiskunde & Informatica (CWI).

Biography He started his career in computer science as research assistant in 1975. As of 1979 he was scientific researcher and lecturer at the Vrije Universiteit, Amsterdam. Until 1985 he worked on database security, database programming languages and he developed a relational DBMS, which became a component of a commercial CASE environment from 1985 to 1991. He was visiting researcher at the University of California, San Francisco (1980 and 1983), visiting researcher Stanford University (2001 and 2002), and Microsoft Research (2005).

In 1985 he moved to Centrum Wiskunde & Informatica (CWI), the national research institute for mathematics and computer science in the Netherlands, to establish the Database Research Group. Between 1986 and 1990, he was co-designer of the PRISMA database machine, an RDBMS for a 100-node multiprocessor. In a follow-up ESPRIT-II project Kersten was responsible for the development of an enhanced version of SQL for documents and geographical data. From 1989 to 1993, he led a national project on the exploitation of the Amoeba distributed system for advanced database management and a national project for database design formalizations.

In April 1992 he became the head of the department of Information Systems. At the same time he started the ESPRIT-III Pythagoras project aimed at performance quality assessment of advanced database systems. He remained associate professor at the Vrije Universiteit, teaching advanced courses on database technology until mid-1994. In 1992 he became an associated professor at the University of Amsterdam and a full professor in multimedia databases in January 1994.

In 2020 he became Knight of the Order of the Netherlands Lion.

Publications

Martin Kersten has published more than 150 publications. In 2009, a team of researchers from the CWI Database Architectures group, composed of Milena Ivanova, Martin Kersten, Niels Nes and Romulo Goncalves, won the "Best Paper Runner Up" at annual ACM SIGMOD conference for their work for on "An Architecture for Recycling Intermediates in a Column-store". In August of the same year, Peter Boncz, Stefan Manegold and Martin Kersten received the VLDB 10-year Best Paper Award for their publication titled "Database Architecture Optimized for the New Bottleneck: Memory Access".

In 2011, Martin Kersten received the VLDB Challenges & Visions Track Best Paper Award for the paper "The Researcher's Guide to the Data Deluge: Querying a Scientific Database in just a Few Seconds", co-authored with his CWI colleagues Stefan Manegold, Stratos Idreos and Erietta Liarou.

On June 26, 2014, at the annual ACM SIGMOD/PODS Conference, Kersten received the ACM SIGMOD Edgar F. Codd Innovations Award for innovative and significant contributions to database systems and databases. Two years later, May 26, 2016, at the annual ACM SIGMOD/PODS Conference, Kersten received the SIGMOD Systems Award for the design and implementation of MonetDB, a pioneering main-memory database system based on a columnar data organization. In the same year, December 8, 2016, Kersten became ACM Fellow.

Other publications include: 1992: PRISMA/DB: a parallel, main memory relational DBMS 1999: Database architecture optimized for the new bottleneck: Memory access 2001: Efficient relational storage and retrieval of XML documents 2002: XMark: A benchmark for XML data management 2007: Database Cracking 2008: Breaking the memory wall in MonetDB 2008: Column-store support for RDF data management: not all swans are white 2009: An Architecture for Recycling Intermediates in a Column-Store 2011: The Data Cyclotron Query Processing Scheme 2011: The Researcher's Guide to the Data Deluge: Querying a Scientific Database in just a Few Seconds 2011: SciBORQ: Scientific data management with Bounds On Runtime and Quality 2012: MonetDB/DataCell: online analytics in a streaming column-store 2012: Data vaults: a symbiosis between database technology and scientific file repositories 2013: Column imprints: a secondary index structure

MonetDB

MonetDB is a high-performance column-store relational database management system with automatic index management, flexible optimizer infrastructure, and programmable backend functionality. MonetDB (initially only called Monet) was first created by 2002 doctoral student Peter Boncz and professor Martin Kersten as part of the 1990s MAGNUM research project at University of Amsterdam. The first version under an open-source software license (a modified version of the Mozilla Public License) was released on September 30, 2004.

MonetDB introduced innovations at all database management system layers: a storage model based on vertical fragmentation, a modern CPU-tuned query execution architecture that often gave MonetDB a speed advantage on the same algorithm over a typical interpreter-based RDBMS. MonetDB is one of the first database systems to focus its query optimization effort on exploiting CPU caches. MonetDB also features automatic and self-tuning indexes, run-time query optimization, and a modular software architecture.

In 2013 Martin Kersten, together with Ying Zhang, Niels Nes and Sjoerd Mullender established a database services CWI spin-off company.