Overview of the Research Group

Welcome at the web pages of the Parallel and Distributed Systems research group (Arbeitsgruppe Parallele und Verteilte Systeme - PVS). The group has been established in October 2010 and is lead by Prof. Dr. Artur Andrzejak. Our research focuses primarily on the following domains.

Dependability of complex software systems. Our research here covers:

  • Detection and isolation of latent software bugs (esp. software aging)
  • Diagnosis of configuration defects
  • Efficiency and robustness of large-scale software testing
  • Performance optimization of code.

Scalable Data Analysis. Topics here include:

  • Scalable data-driven modeling of computer systems
  • Support for analyzing large data sets by non-experts.

In search for solutions we combine approaches from statistics, machine learning, performance evaluation, software engineering and others.

Current Research Projects

Projects related to software dependability include:

  • Detecting of software aging and isolating its root causes in large applications
  • Evaluating test quality and test redundancy in a very large software project (cooperation with Quality Assurance team of SAP SE, Walldorf)
  • Automated identification of failure-inducing changes and misconfigurations in Java applications
  • Identifying performance errors and performance tuning of code.

Current projects related to scalable analysis of data include:

  • Frameworks and algorithms for incrementally-parallel processing and analysis of large-scale data
  • Automated consolidation of data formats from multiple sources (cooperation with a major international consulting company).

Researchers interested in our work and students who would like to participate in these projects (via bachelor/master thesis work or via "Softwarepraktikum") are encouraged to contact us for further details.