In addition to theory and experiment, numerical simulations have become a third pillar for the process of gaining scientific knowledge. In Scientific Computing the tools of numerical mathematics are provided with the aid of computer science to solve questions from a specific scientific subject.
Researchers at the University Oldenburg have access to resources for scientific computing, which are funded by the Deutsche Forschungsgemeinschaft (DFG) and the Federal Ministry for Economic Affairs and Energy (BMWi).
Currently, the IT Services operate HPC Clusters with a total of about 13.500 compute cores and 98 TB of main memory. The total peak performance of the HPC systems is approximately 472 TFlop/s.
There are serveral scientifc applications (e.g. Matlab, R, Stata, etc.) and numeric libraries (e.g. BLAS, Lapack, FFTW, etc.) installed on the Cluster. There are also applications for developing your own programs installed (Compiler, MPI, etc.). Besides a full list of every installed program, you can find much advice about the proper usage of the cluster in the HPC-Wiki.
If you have any problems regarding software installations, the realisation of projects or any other questions, feel free to contact the Scientific Computing via email: hpc38ppsupj6legpoa1rt(at)uni-o1mn8ldefaainbur
76cg.de (firstname.lastname@example.org). Other contact informations can be found in the right column of this website. You can also find the a link to the servicedesk of the IT-Services, which is used to request an HPC-Account, there. You can find a short explanation on how to use this process here.
The Scientific Computing is organizing a meeting every third month. All user reprensentatives are invited to talk about general informations regarding the cluster. A list of the participating user representatives and the protocols can be found in the Wiki.