Teaching

• Supervision of the course on data structures (Kurs 1663 Datenstrukturen), taking place each summer semester.
• Supervision of the course on compiler construction (Kurs 1810 Übersetzerbau), taking place each winter semester.
• Non-regular events. Actually, a seminar on query optimization in database systems scheduled for winter semester 1998/99. (Seminar 01902: Anfrageoptimierung in Datenbanksystemen)

Research

Architectures of modular extensible systems

Examples are the commercially available Informix Universal Server, research prototypes like Predator, or our group's SECONDO project.

 

Interfaces for extension modules

An extensible database system typically consists of a pre-implemented system frame providing well-known database functionality like transaction management, concurrency control, and recovery. These are common features of all database systems, independent of their underlying data model. An extension module is a set of new data types and operators on that types, provided by a database implementor (DBI). It can be registered with the system frame via the implementation of a number of support functions which during query processing will be called by the frame to create, delete, load, and save objects of types defined by the DBI, perform operations on those types, estimate operator execution costs, etc.

Here my focus of research is on the following aspects of extension module implementation:

• Tools for the implementation of extension modules
• Design of an interface between extension modules and the system frame
• Properties of interfaces between cooperating modules

 

Query optimization in modular extensible systems

While a lot of research effort has been put on the investigation of query optimization methods for relational DBMSs, optimizing queries whose operators as well as their argument data types are not predefined is still an open issue.

Publications

• S. Dieker, Efficient Integration of Query Algebra Modules into an Extensible Database Framework, Dissertation, Mensch und Buch Verlag, Berlin 2001, ISBN 3-89820-226-7. Abstract, Full text (pdf, 800 KB), Slides of my defense presentation (pdf, 220 KB, in German).
• S. Dieker, R.H. Güting, and M. Rodriguez Luaces, A Tool for Nesting and Clustering Large Objects. Proc. of the 12th Int. Conf. on Scientific and Statistical Database Management (SSDBM 2000), IEEE, 169-181, July 2000. Abstract, Paper.pdf (technical report) (pdf, 302 KB).
• S. Dieker and R.H. Güting, Plug and Play with Query Algebras: SECONDO. A Generic DBMS Development Environment. FernUniversität Hagen, Informatik-Report 249, February 1999. , Paper.ps.gz (gzip'ed postscript, 59 KB), Paper.pdf (pdf, 123 KB).
Proc. of the Int. Database Engineering and Applications Symposium (IDEAS 2000), 380-390, September 2000.
• S. Dieker and R.H. Güting, Efficient Handling of Tuples with Embedded Large Objects. Data & Knowledge Engineering, 32(3):247-269, March 2000. Abstract, Paper.ps.gz (gzip'ed postscript, 75 KB)
• R.H. Güting, S. Dieker, C. Freundorfer, L. Becker, and H. Schenk, SECONDO/QP: Implementation of a Generic Query Processor. Proc. of the 10th Int. Conf. on Database and Expert System Applications (DEXA'99), LNCS 1677, Springer Verlag, 66-87, 1999. Abstract, Paper.ps.pz (gzip'ed postscript, 46 KB)
, Paper.pdf (pdf, 140 KB), Slides.pdf (pdf, 40KB).