Intel Research Seminar

DATE: Monday July 21, 2003
TIME: Noon - 1:30 pm
PLACE: Intel Seminar (417 S. Craig Street - 3rd Floor)

Hermann Härtig
Dresden University of Technology, Germany

Micro-Kernels - Will they ever deliver?

In the days of the MACH project, many expectations were placed on micro-kernel technology. Later, many of them were disappointed (such as operating system architectures based on multiple generalized servers) or delivered by other technologies(such as multiple os-personalities on single machines). However, combining legacy OSes with other system software that has radically different requirements seems the killer application for micro-kernels. In Dresden, two such areas are investigated: small trusted systems with high assurance requirements and real-time applications, both running besides and reusing L4Linux, a user-level variant of Linux as a legacy OS. We present architectures and some performance results that deny the long-held perception that micro-kernel-based architectures are too expensive for any useful application.

Hermann Härtig is a professor since 1994 at Dresden University of Technology, Germany. His research interests include operating system support for multimedia systems, µ-kernel-based operating systems, real-time operating systems, and security in operating systems. He graduated from the University of Karlsruhe (PhD 1984). From 1984 until 1994 he was the principle developer of the BirliX Operating System at GMD(German National Research Center for Information Technology). He has had extended visits at the International Computer Science Institute in Berkeley, UNSW in Sydney, Hebrew University in Jerusalem, MIT in Boston, and the University of Hildesheim.

For Further Seminar Info:
Contact Kim Kaan, 412-605-1203, or visit

SDI Home: