The hierarchical hypercube network is suitable for massively parallel systems. An appealing property of this network is the low number of connections per processor, which can faci...
The UPIO (User-controllable Parallel I/O) we proposed xtends the abstraction of a linear file model into an n-dimensional file model, making it possible to control the layout of d...
Jang Sun Lee, Sung Hoon Ko, Sanjay Ranka, Byung Eu...
Although platform-independent runtime systems for parallel programming languages are desirable, the need for low-level optimizations usually precludes their existence. This is bec...
A concurrent partitioner for partitioning unstructured finite element meshes on distributed memory architectures is developed. The partitioner uses an element-based partitioning st...
Algorithmic skeletons intend to simplify parallel programming by providing a higher abstraction compared to the usual message passing. Task and data parallel skeletons can be dist...