GPU Computing (CUDA2)

Programming, Programming - other

This course introduces GPU computing on NVIDIA graphics cards, covering GPU hardware, architecture and the design of parallel algorithms. It presents practical use of CUDA, OpenCL and OpenACC, and performance considerations for real applications.

The course is theory-focused, emphasizing GPU procedures, best practices and optimization patterns. Topics include synchronization, matrix multiplication, textures, CUDA-specific features, libraries and profiling. Delivered in Czech or English with partner Sprinx Systems.

Location, current course term

Contact us

Custom

Customized Training (date, location, content, duration)

The course:

Hide detail

• Introduction to CUDA

• CUDA memory types

• CUDA kernels

• CUDA compute capabilities

• Basic techniques and best practices

• Synchronization

• Matrix multiplication

• Textures in CUDA

• CUDA-specific features

• Profiling in CUDA

• Libraries for CUDA

• CUDA and programming languages

Assumed knowledge:

Basic programming skills in a structured programming language.

Recommended previous course:

Introduction to CUDA (CUDA1)

Recommended subsequent course:

GPU Computing - Practical Lab (CUDA3)

Schedule:

2 days (9:00 AM - 5:00 PM )

Education Partner:

Sprinx Systems

Language: