Upcoming new generations of compute and storage resources will comprise heterogenous components to provide the required system performance while maintaining or reducing the energy footprint.
Our research work is focused on the evaluation of accelerator and many-core platforms including GPGPU, FPGA, CellBE and SIMD platforms. Part of the work is the development of enhanced algorithmus and approaches which support these many-core platforms efficiently.
Currently, our targeted application classes include molecular simulations, bioinformatics / life science applications and fault tolerance in large parallel applications.
Thomas Steinke, Kathrin Sobe, Johannes Bock