20 - 22 September, 2017

Introduction

A three-days training workshop on GPU: Graphical Processing Unit/ CUDA: Compute Unified Device Architecture is being organized at NCP from September 20-22, 2017. GPUs have recently emerged as a powerful tool to run the tasks in parallel fashion to reduce the execution time of algorithms by using CUDA language. The course will be helpful for professionals and students who are planning to adopt GPU techniques for parallel computing for their projects. This course will include lectures as well as hands on lab sessions and covers the following topics:

Topics

• GPU computing, CUDA overview, GPU architecture
• CUDA threads, blocks, grids, kernel
• Memory Organization: host, device, register, constant, shared, local and global memories
• Asynchronous operations, optimization (warp, atomics)

Organizing Committee

Workshop Coordinator

Hassan Shahzad (NCP)
Email: hassan.shahzad@ncp.edu.pk
Tel: +92 (51) 2077300 Ext. 322

Workshop Brochure

To download the school brochure, click here.

Worshop Timing

• Worshop timing will be from 0900 to 1230
• Registration will start at 0830 – 0900 on first day

Participation

Applications are invited from students and researchers who are involved in scientific computing in their research projects. PKR. 1000/- is registration fee for participation. On successful completion of workshop certificates will be awarded to the participants. No travel grant will be given to the participants

How to apply

Application form is available at here. and should be submitted to (caad@ncp.edu.pk) not later than 20 August, 2017. List of shortlisted/accepted applicants will be posted at NCP website and also be informed by email once the selection procedure is completed.