Blazing-fast code using GPUs and maore, with Microsoft Visual C++

Daniel Moth

Informations

• Publié par: mtoledan
• Publié le: 20 juin 2011
• Nombre de lectures: 131
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Daniel Moth, Parallel Computing Platform, Microsoft Heterogeneous platform support in Visual Studio  Context  Code  Closing thoughts 146X 36X 19X 17X 100X Interactive Ionic placement for Transcoding HD Simulation in Astrophysics N- visualization of molecular video stream to Matlab using .mex body simulation volumetric white dynamics H.264 file CUDA function matter connectivity simulation on GPU 149X 47X 20X 24X 30X Financial GLAME@lab: An Ultrasound Highly optimized Cmatch exact string simulation of M-script API for medical imaging object oriented matching to find LIBOR model with linear Algebra for cancer molecular similar proteins and swaptions operations on GPU diagnostics dynamics gene sequences  CPU GPU  Low memory bandwidth  High memory bandwidth  Higher power consumption  Lower power consumption  Medium level of parallelism  High level of parallelism  Deep execution pipelines  Shallow execution pipelines  Random accesses  Sequential accesses  Supports general code  Supports data-parallel code  Mainstream programming  Niche/exotic programming  CPUs and GPUs coming closer together…  …nothing settled in this space, things still in motion…  We have designed a mainstream solution not only for today, but also for tomorrow  Part of Visual C++  Visual Studio integration  STL-like library for multidimensional data  Builds on DirectX performance productivity portability  Context  Code  Closing thoughts  void AddArrays(int n, int * pA, int * pB, int * pC) { for (int i=0; i