One of the main challenges in numerical computing on modern high performance clusters for the simulation of real world phenomena is the efficient handling and management of the simulation domain that is usually distributed among computational resources. Classical approaches may store the complete topology with each unit. This entails a significant memory requirement and expensive global communication overhead to ensure consistency when the domain configuration is altered. Our current work addresses this shortcoming by employing a decentral approach to domain organisation, where the essential idea is to limit the domain view of each participating unit to their direct neighbours. The new approach affects various parts of the numerical code significantly. Among them, the octree-based domain generation, which had to be revamped from a central based one to support the decentral structure similarly. To this end, an algorithm has been devised which refines an input geometry on all processes up to a predetermined depth, before distributing the resulting leaf nodes of the geometry to be starting points for a subsequent refinement on their respective processes. We will discuss the key points of the approach, including the distribution of the basis tree, avoiding costly communication for the neighbourhood search by processes refining also their neighbouring nodes and tree balancing in the decentral environment.     «

One of the main challenges in numerical computing on modern high performance clusters for the simulation of real world phenomena is the efficient handling and management of the simulation domain that is usually distributed among computational resources. Classical approaches may store the complete topology with each unit. This entails a significant memory requirement and expensive global communication overhead to ensure consistency when the domain configuration is altered. Our current work addres...     »