Minimize the Gap: How to Avoid Overhead in Database Processing through Co-Design
Jens Teubner, TU Dortmund
Exploiting the potential of modern computing systems requires very careful tuning throughout the system stack. In practice, however, a lot of the available potential gets lost behind abstractions, some of which are decades old.

In the talk, I will show how better abstractions and deliberate co-design between system layers can recover some of those losses. Specifically, I will emphasize on “Flounder IR” as a new interface language to generate JIT-compiled code at unprecedented speeds. And I will present “MxKernel”, our vision of a new way of architecting the interplay between operating systems and database systems.

The work I present was done jointly with Henning Funke and Jan Mühlig from TU Dortmund University and funded by DFG (Deutsche Forschungsgemeinschaft).

Please email for a Zoom link
About the speaker
Jens Teubner is the head of the Databases and Information Systems Group at TU Dortmund University since April 2013. Previously, he held positions as a postdoc at the IBM T.J. Watson Research Lab in New York, USA and as a senior researcher ("Oberassistent") at the Systems Group of ETH Zürich. He holds a PhD in Computer Science from TU München (graduated in 2006 under the guidance of Torsten Grust; topic: XML query processing) and a Masters degree ("Diplom") in physics from the University of Konstanz in Germany (advisor: Ernst Bucher).

The focus of Jens Teubner's research is the implementation of database systems on modern hardware architectures. He is the founder of Avalanche, a research project at ETH Zürich that has quickly become an international leader in the use of field-programmable gate arrays (FPGAs) for database tasks. He co-developed one of the fastest in-memory join implementations available today and showed how hardware acceleration in modern network cards (by means of Remote Direct Memory Access, RDMA) can be used to speed up join processing in distributed databases. During his PhD, Jens Teubner was concerned with scalable evaluation techniques for XML, in particular by using relational database technology as a back-end. The outcome of this work—the XQuery compiler Pathfinder—has become known as part of the open source system MonetDB/XQuery.
Date & Time
Thursday, April 22, 2021 - 14:00