Partnering with NVIDIA, this professional four day course is designed for programmers who are looking to develop comprehensive skills in writing and optimizing applications that fully leverage multi-core processing capabilities of GPUs.
Your fee includes
- Use of a laptop equipped with CUDA capable GPU
- Choice of Windows and Linux OS
- Printed manual of all lectures
- Electronic copy of lab exercises
- Certificate of Completion
- Beverages and snacks
Space is limited - Please register early to guarantee your spot
Your Instructor
|
Kelly Goss - Training Program Manager
Kelly is the Training Program Manager for Acceleware. She has taught over 25 parallel programming courses for students from a diverse range of industries and backgrounds. Kelly is also a member of Acceleware’s development team, assisting in the writing and optimization of Acceleware applications and providing consulting services to our clients. Kelly has a PhD in Electrical Engineering from the University of Calgary.
|
Schedule
Tue-Fri: 9:00AM – 5:00PM (includes a 1 hour break for lunch)
Agenda
- Day 1:
- Overview of GPU computing
- Data-parallel architectures and the GPU programming model
- GPU memory model & thread cooperation
- Hands-on exercises: GPU memory management, simple CUDA kernels and shared memory and constant memory
- Day 2:
- Asynchronous operations
- Advanced CUDA features
- Libraries
- Debugging GPU Programs
- Hands-on-exercises: Asynchronous operations, CUDA features, experience with CUFFT, CUBLAS, Thrust, debugging
- Day 3:
- Introduction to optimization
- Resource management, latency and occupancy
- Memory performance optimizations
- Profiling applications
- Hands-on exercises: Arithmetic optimizations, occupancy calculator, profiling and memory access patterns
- Day 4:
- CUDA compiler and user-defined libraries
- OpenACC
- Hands-on exercises: Case study exercise and OpenACC
- Case study: Finite difference stencil algorithm or monte carlo simulations
|
All lectures are a combination of teaching and hands-on tutorials
NVIDIA’s foundational training material is augmented with Acceleware’s experience over the past 8 years and with examples specific to an HPC audience
