Get in Touch
 Duration 28 hours

Course Outline

Introduction

  • Fundamentals of CUDA.
  • Comparing CUDA with OpenCL and SYCL.
  • An overview of CUDA’s key features and architecture.
  • Configuring the development environment.

Getting Started

  • Creating a new CUDA project in Visual Studio Code.
  • Examining the project structure and associated files.
  • Compiling and executing the program.
  • Displaying output using printf and fprintf functions.

CUDA API

  • Understanding the function of the CUDA API within host programs.
  • Querying device information and capabilities via the CUDA API.
  • Managing device memory allocation and deallocation using the CUDA API.
  • Transferring data between host and device using the CUDA API.
  • Launching kernels and synchronizing threads via the CUDA API.
  • Handling errors and exceptions using the CUDA API.

CUDA C/C++

  • Understanding the role of CUDA C/C++ in device programs.
  • Writing CUDA C/C++ kernels for GPU execution and data manipulation.
  • Utilizing CUDA C/C++ data types, qualifiers, operators, and expressions.
  • Applying CUDA C/C++ built-in functions for math, atomics, and warp operations.
  • Using CUDA C/C++ built-in variables such as threadIdx, blockIdx, and blockDim.
  • Leveraging CUDA C/C++ libraries like cuBLAS, cuFFT, and cuRAND.

CUDA Memory Model

  • Distinguishing between host and device memory models.
  • Working with CUDA memory spaces: global, shared, constant, and local.
  • Managing CUDA memory objects, including pointers, arrays, textures, and surfaces.
  • Applying CUDA memory access modes such as read-only, write-only, and read-write.
  • Understanding CUDA memory consistency and synchronization mechanisms.

CUDA Execution Model

  • Comparing host and device execution models.
  • Defining parallelism using CUDA threads, blocks, and grids.
  • Utilizing CUDA thread functions like threadIdx, blockIdx, and blockDim.
  • Applying CUDA block functions such as __syncthreads and __threadfence_block.
  • Working with CUDA grid functions, including gridDim, gridSync, and cooperative groups.

Debugging

  • Identifying common errors and bugs in CUDA applications.
  • Using the Visual Studio Code debugger to inspect variables, breakpoints, and call stacks.
  • Debugging CUDA programs on Linux using CUDA-GDB.
  • Detecting memory errors and leaks with CUDA-MEMCHECK.
  • Analyzing and debugging CUDA programs on Windows using NVIDIA Nsight.

Optimization

  • Understanding factors influencing CUDA program performance.
  • Improving memory throughput using CUDA coalescing techniques.
  • Reducing memory latency through CUDA caching and prefetching.
  • Optimizing memory access and bandwidth using CUDA shared and local memory strategies.
  • Measuring and enhancing execution time and resource utilization via CUDA profiling tools.

Requirements

  • A solid understanding of the C/C++ language and fundamental parallel programming concepts.
  • Basic knowledge of computer architecture and memory hierarchy.
  • Practical experience with command-line interfaces and code editors.

Target Audience

  • Developers seeking to master CUDA programming for NVIDIA GPUs to harness their parallel power.
  • Developers aiming to create high-performance, scalable code compatible with various CUDA-enabled devices.
  • Programmers interested in exploring the low-level intricacies of GPU programming to optimize code performance.

Number of participants


Price per participant

Upcoming Courses

Related Categories