Заседание семинара "Высокопроизводительные вычисления" 14.09.2017 в 10-00. "Two reports: 1. Structure and Research Activities of the Institute of Information and Communication Technologies, Bulgarian Academy of Sciences; 2. On the parallel implementation of quasi-Monte Carlo on Intel MIC Architecture.", Todor Gurov, Aneta Karaivanova
The recent advances in HPC increasingly rely on the use of accelerators and other similar devices that improve the energy efficiency and offer better performance for certain type of computations. The Xeon Phi co-processors combine efficient vector floating point computations with familiar operational and development environment.
In order to allow the quasi-Monte Carlo algorithms to make use of hybrid OpenMP+MPI programming, we implemented generation routines that save both memory space and memory bandwidth, with the aim to widen the applicability of quasi-Monte Carlo algorithms in environments with an extremely large number of computational elements. We present our implementation and compare it with regular Monte Carlo using a popular pseudorandom number generator, demonstrating the applicability and advantages of our approach. We present also adapted quasi-Monte Carlo algorithms for numerical integration and matrix algorithms, and their timing and scalability results.