LONI | Training | Workshops | Nvidia GPU Workshop

Nvidia GPU Workshop

Title	Date	Times	Venue
Nvidia GPU Workshop	June 1 - 2 2016	800 AM - 500 PM	BEC ROOM 1615

Event Details

NVIDIA and LSU Research Computing and HPC are pleased to be organizing a 2-day High Performance Computing and Programming event on June 1-2, 2016 in Business Education Complex, Room 1615 at LSU, Baton Rouge, LA.

Why you should attend

NVIDIA GPUs are the world’s fastest and most efficient accelerators delivering world record scientific application performance. NVIDIA’s CUDA Technology is the most pervasive parallel computing model, used by over 250 scientific applications and over 150,000 developers worldwide. This Programming Workshop will focus on introducing scientific computing programming utilizing NVIDIA GPUs to accelerate applications across a diverse set of domains.

Presented by NVIDIA instructor Dr. Jonathan Bentz, the workshop will introduce programming techniques using CUDA and OpenACC paradigms as well as optimization, profiling and debugging methods for GPU programming. An introduction to Deep Learning using GPUs will also be covered.

Who it's for

Undergraduate/Graduate Students, Postdocs, Researchers, and Professors

Agenda

June 1st | Day 1

Time	Topic
9:00 AM - 4:30 PM	Introduction to GPU programming High Level Overview of GPU architecture OpenACC: An introduction on compiler directives to specify loops and regions of code in standard C, C++ and Fortran to be offloaded from a host CPU to an attached accelerator Hands-On examples to focus on data locality GPU-Accelerated Libraries: discussion including AmgX, cuSolver, cuBLAS and cuDNN Basics of GPU Programming; An introduction to the CUDA C/C++ Language 4 Hands-On examples will Illustrate simple kernel launches and using threads

Time

Topic

9:00 AM - 4:30 PM

Introduction to GPU programming

High Level Overview of GPU architecture
OpenACC: An introduction on compiler directives to specify loops and regions of code in standard C, C++ and Fortran to be offloaded from a host CPU to an attached accelerator
Hands-On examples to focus on data locality
GPU-Accelerated Libraries: discussion including AmgX, cuSolver, cuBLAS and cuDNN
Basics of GPU Programming; An introduction to the CUDA C/C++ Language
4 Hands-On examples will Illustrate simple kernel launches and using threads

June 2nd | Day 2

Time	Topic
9:00 AM - 4:30 PM	Performance and Optimization Overview of Global and Shared memory usage Hands-On examples will illustrate a 1D Stencil and Matrix Transpose Using NVIDIA Profiler to identify performance bottlenecks Advanced Optimizations using Streams and Concurrency to overlap communication and computation Hands-On examples will use CUBLAS with Matrix Multiply Conclude with a Deep Learning Overview Intro to Deep Learning / Machine Learning Overview of Global and Shared memory usage (Caffe, Torch, Theano) Deep Learning with GPUs and NVIDIA DIGITS Live demo using NVIDIA DIGITS Caffe Lab Examples (time permitting) Deep Learning is a rapidly growing segment born from the fields of artificial intelligence and machine learning. It is increasingly used to deliver near-human level accuracy for image classification, voice recognition, natural language processing, sentiment analysis, recommendation engines, and more. Applications areas include facial recognition, scene detection, advanced medical and pharmaceutical research, and autonomous, self-driving vehicles. This overview will focus on introducing attendees to the use of GPU accelerated deep learning frameworks utilizing NVIDIA GPUs for ideal performance and scalability.

Time

Topic

9:00 AM - 4:30 PM

Performance and Optimization

Overview of Global and Shared memory usage
Hands-On examples will illustrate a 1D Stencil and Matrix Transpose
Using NVIDIA Profiler to identify performance bottlenecks
Advanced Optimizations using Streams and Concurrency to overlap communication and computation
Hands-On examples will use CUBLAS with Matrix Multiply
Conclude with a Deep Learning Overview

Intro to Deep Learning / Machine Learning

Overview of Global and Shared memory usage (Caffe, Torch, Theano)
Deep Learning with GPUs and NVIDIA DIGITS
Live demo using NVIDIA DIGITS

Caffe Lab Examples (time permitting)

Deep Learning is a rapidly growing segment born from the fields of artificial intelligence and machine learning. It is increasingly used to deliver near-human level accuracy for image classification, voice recognition, natural language processing, sentiment analysis, recommendation engines, and more. Applications areas include facial recognition, scene detection, advanced medical and pharmaceutical research, and autonomous, self-driving vehicles. This overview will focus on introducing attendees to the use of GPU accelerated deep learning frameworks utilizing NVIDIA GPUs for ideal performance and scalability.

Coffee and Lunch will be provided.

About the Instructor

Dr. Jonathan Bentz is a Solution Architect with NVIDIA, focusing on Higher Education and Reasearch customers. In this role he works as a technical resource to customers and OEMs to support and enable adoption of GPU computing. He delivers GPU training such as workshops to train users and help raise awareness of GPU computing. He also works with ISV and customer applications to assist in optimization for GPUs through the use of benchmarking and targeted code development efforts. Prior to NVIDIA, Jonathan worked for Cray as a software engineer where he developed and optimizaed high performance scientific libraries such as BLAS, LAPACK, and FFT specifically for the Cray platform. Jonathan obtained his PhD in physical chemistry and his MS in computer science from Iowa State University.

Important:

Please bring your laptop to participate in hands-on exercises. A GPU in your laptop is not required. No previous GPU programming experience is required. However, beginner-level C and Linux experience will be expected.

In preparation, you may watch a few short (approx. 5 min) YouTube videos on introductory GPU programming topics. These CUDACasts can be found at this link. Also, please take time to register at CUDA developer site: link here.

Registration

Please register through this link.