Large memories have become an affordable storage medium for databases involving hundreds of Gigabytes on multi-processor systems. In this short note, we review our research on building relational engines to exploit this major shift in hardware perspective. It illustrates that key design issues related to parallelism pose architectural problems at all levels of a system architecture and whose impact is not easily predictable. The sheer size/complexity of a relational DBMS and the sliding requirements of frontier applications are indicative that a substantial research agenda remains wide open.

Springer
Lecture Notes in Computer Science
European Conference on Parallel Processing
Database Architectures

Kersten, M., Manegold, S., Boncz, P., & Nes, N. (2001). Macro- and Micro- Parallelism in a DBMS. In Proceedings of European Conference on Parallel Processing 2001 (EuroPar 0) (pp. 6–15). Springer.