CUDA & OpenACC Workshop 2014

CUDA & OpenACC Workshop 2014

By NVIDIA Deep Learning Institute

Date and time

Friday, July 11, 2014 · 9am - 5pm PDT

Location

Creation Theatre at Matrix, Biopolis

Description

The CUDA & OpenACC Workshop 2014 will provide an overview of the three main vectors to leverage the computing power of GPU: CUDA programming language, accelerated libraries and OpenACC directives driven programming.

Attendees will be given basic introduction to these 3 topics in form of 1 hour lectures. Every course will be followed by a 1 hour hands-on session where participants will have the chance to test the new notions.

The last session aims to introduce the new Tegra K1 fully programmable embedded chip and deliver a hands-on experience using a Jetson K1 board which allows to unleash the power of 192 CUDA cores to develop solutions in computer vision, robotics, medicine, security, and automotive.

PREREQUISITES

The attendees will needs to have some basic C/C++ knowledge. Familiarity in Linux is preferred though not mandatory as a remote connection on a CUDA enabled server will be used to try out programming exercises.

HARDWARE

Every attendant will be required to use its personal laptop for the hands-on. The hardware specifications DOES NOT need to include an NVIDIA GPU. The laptop should support a wireless Wi-Fi connection which will be provided free of charge.

LANGUAGE

The course will be entirely held in English.

OUTCOME

By the end of this workshop attendees will have built a number of accelerated applications and have a basic understanding of the GPU architecture as well as different methods to leverage its parallelism. Knowledge acquired is general enough to be applied on any application in any field. Being an introductory course a lot of advanced features are outside the scope of the workshop although references for further reading will be provided.

AGENDA

9am-10am

GPU architecture overview and CUDA programming language introduction (lecture)

10am-11am

GPU architecture overview and CUDA programming language introduction (hands-on)

11am-12pm

Drop-in acceleration using CUDA libraries (lecture)

12pm-1pm

Lunch break

1pm-2pm

Drop-in acceleration using CUDA libraries (hands-on)

2pm-3pm

Directives based programming: OpenACC (lecture)

3pm-4pm

Directives based programming: OpenACC (hands-on)

4pm-5pm

Acceleration in embedded form factor: Jetson K1 development board hands-on

Registration is free and By Invitation Only. Please RSVP online or send your registration to Gabriel Noaje at g.noaje@adsc.com.sg

Organized by

Sales Ended