We show how generative and meta-programming techniques can be applied to the implementation of a skeleton-based parallel programming library . The resulting library, QUAFF, both offers a high level of abstraction and produces high performance code by performing most of the high to low-level transformations at compile-time rather than run-time. The implementation is derived directly from a set of explicit production rules, in a semantic-oriented style, it is therefore formally sounded and much more amenable to proofs or extensions
Keywords for this software
References in zbMATH (referenced in 3 articles , 1 standard article )
Showing results 1 to 3 of 3.
- Cordier, S.; Coullon, H.; Delestre, O.; Laguerre, C.; Le, M.H.; Pierre, D.; Sadaka, G.: FullSWOF$\_$Paral: comparison of two parallelization strategies (MPI and SKELGIS) on a software designed for hydrology applications (2013)
- Fraguela, Basilio B.; Bikshandi, Ganesh; Guo, Jia; Garzarán, María J.; Padua, David; Von Praun, Christoph: Optimization techniques for efficient HTA programs (2012)
- Falcou, Joel; Sérot, Jocelyn: Formal semantics applied to the implementation of a skeleton-based parallel programming library (2008)