Hallo C++ Freunde,
ich habe den Video-Mitschnitt der Vortrags von Erik Zenker gestern zum nachsehen hochgeladen.
Vielen Dank nochmal an Erik für den coolen Vortrag.
Ich wünsche Euch ein tolles Wochenende
Andreas
The Last Programming Model for Parallel Kernel Acceleration You Will Ever Need
This talk will dive deep into the library for parallel kernel acceleration Alpaka that provide an uniform abstract C++ interface to a range of parallel programming models such as CUDA and OpenMP. We will show how Alpaka can achieve platform and performance portability across various types of architectures by exploiting parallelism and memory hierarchies at all levels available in current hardware.
Alpaka has demonstrated platform and performance portability in academic (daxpy, dgemm) and real world applications (PIConGPU, HASEonGPU) on various architectures such as CPUs and GPUs. However, the Alpaka C++ template interface allows for straightforward extension of the library to support other and upcoming architectures and specialization of its internals for optimization.