GPU Bootcamp at IUCAA 2019
GPU BOOTCAMP AT IUCAA
Tuesday, April 2 to Wednesday, April 3, 2019
Bhaskara I, IUCAAPune 411007, India
GPU Bootcamp is an exciting and unique way for scientists and researchers to learn the skills needed to start quickly accelerating codes on GPUs. This two-day event will introduce you to available GPU libraries, programming models, and platforms where you will learn the basics of GPU programming. You will also experience extensive hands-on collaboration based on a real-life code using the OpenACC programming model, learn CUDA fundamentals and discuss porting strategies of in-house codes. GPU Bootcamp is organized collaboratively by OpenACC.org, Amazon Web Services, NVIDIA, and Linux Academy.
Agenda
Day 1: Tuesday, April 2, 2019
- Introduction to Heterogeneous Computing
- Why Hybrid Architecture?
- Heterogeneous Programming Models
- Moore’s Law, Dennard Scaling, Amdahl’s Law
- CPU vs. GPU Architecture
- Fundamental Difference Between the Throughput-based vs. Latency-based Architecture
- Ways to GPGPU Programming
- Libraries vs. Directive-based vs. Languages
- Hands-on Directive-based Programming
- Introduction to OpenACC
- Directive-based Programming: Hands-on
- Mini-Application
- Hands-on Session: Real-world Mini-application
Day 2:Wednesday, April 3, 2019
- IUCAA In-House Code
- Discuss Strategy for Porting In-house Applications
- CUDA Languages
- Porting Under Guidance of Mentor
- Introduction to CUDA C API
- Introduction to CUDA Memory Hierarchy
- Hands-on CUDA Programming
Moderator
BHARATKUMAR SHARMA
Senior Solution Architect, NVIDIA
Bharatkumar Sharma is a senior solutions architect at NVIDIA in south Asia. He has about 10 years of research and development experience in software architecture and distributed and parallel computing. He has published papers and journal articles in fields of parallel computing and software architecture. He obtained his master's degree in information technology from the Indian Institute of Information Technology, Bangalore.