Objectives

OpenACC is a directive-based programming model to create portable GPU- accelerated applications. After a short introduction to GPU Programming with CUDA, this course introduces you to the OpenACC standard with an overview of CAPS Compilers extra features that will make you able to improve your OpenACC applications.

Agenda

Day 1

Morning – CUDA Basics

• Introduction to GPU computing
• Nvidia GPU architectures
• CUDA programming model and API

Afternoon – OpenACC (1/2)

• OpenACC overview and compilers
• OpenACC programming model
• Managing data with OpenACC

Day 2

Morning – OpenACC (2/2)

• OpenACC loop constructs
• Asynchronism with OpenACC
• OpenACC runtime APi

Afternoon – OpenHMPP Introduction

• What is in OpenHMPP and not in OpenACC
• OpenHMPP compilation model
• Driving the code generation and gridification
• Automatic unroll, split, jam etc. transformations
• Computing on multiple GPUs
• OpenHMPP API for C, C++ and Fortran

Prerequisites

Knowledge in C

Audience

Developers, project leaders.

OS

Linux

For further information ,please contact sales@caps-entreprise.com