Parallel Algorithms in STAPL: Implementation and Evaluationy

The Standard Template Adaptive Parallel Library (stapl) is a high-productivity parallel programming framework that extends C++ and the STL with unied support for shared and distributed memory paral-lelism. stapl provides distributed data structures (pContainers) and parallel algorithms (pAlgorithms) and a generic methodology for extending them to provide customized functionality. In this paper, we present the design and implementation of a stapl pAlgorithm, p unique copy. We also present preliminary experimental results for p unique copy and another pAlgorithm, p copy if, for comparison. The results illustrate the scalability of the algorithms, as well as the sensitivity of their perfor-mance to the underlying data distribution. Finally, we present initial results based on the usage of stapl anywhere inserter, a construct for parallel insertions that allows for performance optimization when it is used in pAlgorithms