Programming Models

The programming models research sits  in between the life sciences applications and algorithm development on the one hand, and the hardware on the other hand. The aim of our work is to offer the software and runtime infrastructure necessary to implement and execute novel life sciences applications on upcoming HPC systems consisting of heterogeneous compute nodes. Of particular interest to life sciences computing in general are programmer productivity and big data. Combined with the continued need for fine-grained parallelization, which is driven from continuously increasing core counts in hardware, substantial improvements in software and runtime infrastructure are required to make high-performance computing practical for tackling life science problems. We aim to produce a flow programming model and execution infrastructure for writing life sciences applications that can be executed on one or more heterogeneous compute nodes. Continuous work on proto-apps will be used to ensure, demonstrate and measure that these innovations are of value to the project partners.


Comments are closed.