Computing has become a central component of modern scientific research: Large volumes of data are generated by increasingly automated measuring devices, which need to be stored, organized and analysed to extract new insights. New discoveries often only reveal their relevance when compared in context with the already available knowledge, requiring extensive comparison with large public datasets. Computational simulation has become a third pillar of science - after experiments and theory - which allows researchers to challenge their understanding of complex systems in silico. We support scientists in their work by maintaining high-performance computing (HPC) infrastructures, providing direct support to researchers in using these resources and training scientists in the efficient application of scientific high-performance computing.
- Operation of a professional storage infrastructure for scientific data, which is required for managing the vast amounts of data produced in modern life sciences.
- Development and operation of a state-of-the-art HPC infrastructure environment to support large-scale data analysis, modelling and simulation applications.
- Support to research groups and projects on complex data management questions, from requirement analysis to development, deployment and operation of productive software solutions.
- Consulting for researchers on bioinformatics and computational biology aspects of their projects.
- Support to users in computational aspects of project definition for applications to national and international grant programs.
- Organization of training and teaching courses in scientific HPC, bioinformatics, biostatistics and data visualization.
- Operation of computational scientific applications as professional and robust services for the life science community.
More information about our services
We maintain a large suite of software tools for use on two high-performance computing environments that are tailored to specific scientific needs:
This cluster is designed for data and I/O intensive applications (e.g. in life sciences). The infrastructure is designed to meet the requirements of typical Bioinformatics applications and is continuously updated to match the growing needs in life sciences. The current computing power is provided by 130 nodes consisting of over 3,700 cores. A high-performance cluster file system (GPFS) with Infiniband allows the management of petabytes of data and billions of files, and delivers the bandwidth needed by data intensive applications.
The Maia cluster environment is tailored for parallel compute-intensive applications requiring fast interprocess communication (e.g. molecular modelling). The network includes, in addition to the 1GB Ethernet network, an Infiniband network to achieve high-bandwidth for the node interconnection and for the file system services. This cluster was extended over the years and has grown to 290 nodes today with a peak performance of over 55 TFlops.
More information about our resources