As a result of my research work at different institutions, I have a history of running and finalized projects. Here is a short overview.

Active Projects

There is an official list of projects in my group at TU Chemnitz. Beside that, there is a couple of side projects I am working on.

Everything that relates to code is available on GitHub.


OpenSubmit is a small web application for managing student assignment solutions in a university environment.

OpenSubmit offers a trivial web page were students can login and submit their assignment solutions. Teachers and their personal use the backend interface to manage assignments, deadlines, and the gradings. Students are informed about the progress of their correction and their final grade via eMail and the frontend page.

The unique capability of OpenSubmit is the support for coding assignments, were students upload their programming exercise solution as source code archive. OpenSubmit offers an executor daemon that runs on another machine and downloads submitted solutions from the web server. These archives are unpacked and compiled, so that non-compiling assignment solutions are rejected by the system before the deadline. You can also run an assignment-specific validation script that figures out if the student code behaves nicely, before accepting it as solution. This makes the life of the corrector less miserable, because after the deadline, all gradable solutions are ‘valid’. Students also seem to like the idea of having a validated solution, so that they do not fail due to technical difficulties at the correctors side.


The fundamental principle of reliability modeling is the combination of combinatorial analysis and probabilistic methods. Based on an estimated reliability for single components, both analytical and simulation methods investigate the reliability of a system built from these components.

The collection of reliability data is a crucial issue in practice. Both the ever-increasing hardware complexity and the exploding software complexity render the gathering of solid reliability information a great challenge.

Our FuzzTrees approach allow to perform fault tree modeling with imprecise knowledge about component reliability. Furthermore, they enable the reliability engineer to express configurable system variation points directly in the model and its graphical representation.

As part of the research activities, I am maintaining a web-based editor for dependability modeling. Please check the free public version at

Informatik ohne Stecker

This project is a port of Tim Bell's fantastic Computer Science Unplugged program for German kids. The material is freely available for schools and other educational institutions.

All material is available at

Distributed Resource Management Application API

The 'Distributed Resource Management Application API (DRMAA)' working group at the Open Grid Forum (OGF) develops and maintains a set of API specifications for tightly coupled and portable programmatic access to cluster, grid, and cloud systems.

The DRMAA working group deliverables are intended to facilitate the development of portable application programs and high-level libraries such as SAGA or OGSA-BES. DRM system vendors can provide a standardized access to their product through a DRMAA implementation. High-level API designers, meta-scheduler architects and end users can rely on such DRMAA implementations for a unified access to execution resources. The scope of the API standardization is focused on job submission, job control, reservation management, and retrieval of job and machine monitoring information.

DRMAA has it's home at

Finished Projects


CiteMaster was a private project that driven by the needs of my university job. Felix Salfner and me developed a web-based management solution for bibliographies. We implemented automated literature list generation, web page export, collaborative project management, the most detailed data model available in this area, and a lot more ...

The project is no longer maintained, but still runs and serves different web pages (such as this one). The sources are on GitHub, feel free to give it a try at

SINA - Security in Networked Automotive Systems

This was a cooperation project with Audi, targeting novel security analysis ideas for networked automotive systems. We presented some results on the SAE World Congress 2014.

Single Chip Cloud Computer, with Intel Labs Braunschweig (2010-2013)

Intel Labs has created an experimental processor named Single-chip Cloud Computer (SCC). SCC is a research microprocessor that contains 24 tiles with two P90 IA cores per tile . All tiles are connected by a new on-chip messaging network. The chip design furthermore contains advanced power management technologies and 4 integrated DDR3 memory controllers.

Our group is one out of 7 German universities that host a SCC prototype for operating system and firmware research. Our project aims at the establishment of a reliable SMP Linux operating system on the SCC, in order to investigate the challenges for operating system and hypervisor with such a processor architecture.

Early Anomaly Detection in SAP Business ByDesign Systems (2011-2014)

Please contact me directly for further information.

RAS in System/z, with IBM Labs Böblingen (2010)

In cooperation with IBM Labs Böblingen, we worked on new failure prediction and anomaly detection approaches in a System/Z environment.

DAAD PPP program "Acções Integradas Luso-Alemãs" (2011-2012)

This was a small, DAAD funded, research exchange program with our colleagues from Portugal.

International Teaching Professionals

The program "International Teaching Professionals" offers international junior scientists the chance to further their university didactic skills and to network with one another. I acted as mentor in this program.

EFInject - Software-Implemented Fault Injection on Firmware Level (2010-2011)

Software-implemented fault injection (SWIFI) is an established method to emulate hardware faults in computer systems. Existing approaches either extend the operating system by special drivers, modify the runtime environment, or change the application under test.

In cooperation with Fujitsu Technology Solutions in Paderborn, we worked on the realization of a novel fault injection concept based on EFI firmware technology. Our approach extends the firmware of these systems in order to a) make fault injection completely transparent to the operating system, and to b) support a larger variety of fault types. No modifications to the operating system or the application code are needed.

Due to technical limitations of the EFI implementations available today, the project was suspended. Our final software stack is freely available for further research.

European FP7 Integration Project "Plug and Produce Joint Interface Modules (PoPJIM)" (2010)

I helped in the application phase of this project, but left Humboldt University after the project has started. We gathered a grant of 400k€ to work on P2P infrastructures for industry robotics.

Grid Occam (2004)

The Grid-Occam project tried to re-establish the old-fashioned transputer programming language Occam for a grid computing environment. It was funded by Microsoft Research with 25k€. We developed a reasonably well working Grid-Occam compiler (.NET and Java back-end) and a first MPI runtime library. (...more...)

BB-Grid, with TU Berlin, BTU Cottbus and University of Potsdam (2005-2007)

BB-Grid was an initiative from 2005 to 2007, were we connected computational resources of HPI, TU-Berlin, BTU Cottbus and the University of Potsdam into a Globus 3 compute grid.

Adaptive Services Grid (ASG) Infrastructure, with Daimler Chrysler Research (2004-2007)

From 2004-2007, I was the lead of the Services Infrastructure work component in the Adaptive Services Grid EU project, working with a grant of 400k€ for the project duration. The project developed an open platform for the automated and adaptive creation of complex service workflows from semantic service descriptions. Our group at HPI created the Services Infrastructure (SI) layer of this platform. We continuing the research on this service infrastructure under the new label Adaptive eXecution Platform (AXP).

AXP provides an uniform, WSRF-based access to stateful service instances on dynamically allocated execution resources. This includes functionalities for deployment, instantiation, invocation, and monitoring of internal or proxy services. Our current implementation is based on a combination of commercial-of-the-shelf (COTS) middleware products with established standards from both the Web and the grid service community. It supports dynamic resource allocation and service placement for heterogeneous component technologies like .NET and J2EE.

The underlying system model and the implementations acted as central material for my doctoral thesis. Since the thesis was written in German language, it might be easier to get all relevant details from the according papers and the Semantic Service Provisioning book.

SOA Runtime Governance, with Software AG (2006-2008)

We developed a pretty smart solution for enforcing resource partitioning configurations in a distributed application server setup. The whole prototype was realized as add-on for the Software AG CentraSite governance solution. I don't know if any part of our research results finally went into the product ...

Condor, with University of Wisconsin-Madison (2005-2008)

For several years, I was the primary maintainer for one of the third-party Condor Debian packages. Condor is a traditional cluster system for high throughput computing in local and wide-area distributed environments. The Condor people meanwhile provide Debian packages on their own.

As second activity, I helped to complete the DRMAA C library for Condor. The remaining of this project are available at SourceForge.

Rotor, with Microsoft Research

Rotor was a shared source version of the first Microsoft .NET versions. We worked on a project to implement transparent object migration for distributed .NET environments.