The positions were created when Prof. Andrew S. Tanenbaum received a 5-year 2.5 million euro European Research Council “Advanced Grant” to do research in reliable and secure operating systems. Here is a summary of the grant proposal. The basic idea of the proposal is that current operating systems are much too big and very badly structured and the situation is getting worse every year. In this research we want to investigate much smaller, modular, fault-tolerant, multiserver designs. The research to be done by the postdoc will be determined in consultation with Prof. Tanenbaum, but the focus must be on making systems much more reliable and secure than they are now and it should fit into the grant proposal in some way. Funding is also available for several programmers and Ph.D. students; these positions have already been filled.
…
I am proposing replacing the entire monolithic kernel paradigm with a new one consisting of small, tightly constrained modules, running in user mode and each strictly obeying the POLA. Nothing like this has ever been achieved before. I also have many additional goals, such as being able to restart or even replace faulty components on the fly during execution and assign legal liability when faults occur, as described in the full proposal. While my previous work has taken a few baby steps in this direction, there is much more research needed to make this approach viable.
Fortunately, this can be done without affecting the user software by simply having the new operating system emulate the existing and long-stable POSIX interface (the UNIX standard), possibly extended somewhat (e.g.,with some of the Linux system calls). As long as the usual POSIX system calls such as OPEN, READ, EXEC, and so on are available, most UNIX software can be made to work without too much effort.
…
If this work succeeds, it will open up new areas to explore in terms of componentized, multiserver operating systems. People will then examine other kinds of new software components and how they fit in.
The project budget is just under 2.5 M€ and described in the full proposal