With the increasing cost of power in computing and the increasing size of resources, programming approaches need to adapt to allow for optimal utilization of the available resources and account for hard- and software failures, while having high usability. In this regard, the project TaLPas (Task-basierte Lastverteilung und Auto-Tuning in der Partikelsimulation) targets an auto-tuning, task-based approach to high-performance particle simulations. The main goal of TaLPas is to provide a solution to fast, robust and efficient simulation of many, potentially interdependent particle systems in a distributed environment.
|Funding Agency||Federal Ministry of Education and Research|
|Runtime||January 2017 – December 2019|
Scalable Computing & Optimization Methods
Dr. Colin W. Glass