ROCm for Windows Training Course

ROCm is an open-source platform designed for GPU programming, offering robust support for AMD graphics cards alongside compatibility with CUDA and OpenCL standards. By providing direct access to hardware specifics, ROCm grants developers complete authority over the parallelization process. This level of control, however, necessitates a solid grasp of device architecture, memory structures, execution models, and optimization strategies.

The introduction of ROCm for Windows marks a significant development, enabling users to deploy and utilize this platform on the Windows operating system. Widely adopted for both personal and professional workflows, this integration allows users to harness the computational power of AMD GPUs for diverse applications, including artificial intelligence, gaming, graphics processing, and scientific computing.

This instructor-led live training (available online or onsite) is tailored for developers ranging from beginner to intermediate levels who aim to implement ROCm on Windows to program AMD GPUs and fully exploit their parallel capabilities.

Upon completion of this course, participants will be equipped to:

• Establish a development environment on Windows featuring the ROCm Platform, an AMD GPU, and Visual Studio Code.
• Develop a fundamental ROCm program that executes vector addition on the GPU and retrieves results from GPU memory.
• Utilize the ROCm API to access device information, manage device memory allocation and deallocation, transfer data between host and device, initiate kernels, and synchronize threads.
• Employ the HIP language to author kernels that run on the GPU and process data.
• Leverage HIP built-in functions, variables, and libraries to execute standard tasks and operations.
• Optimize data transfer and memory access patterns by utilizing ROCm and HIP memory spaces, including global, shared, constant, and local spaces.
• Apply ROCm and HIP execution models to manage the threads, blocks, and grids that structure parallelism.
• Debug and test ROCm and HIP applications using specialized tools such as the ROCm Debugger and ROCm Profiler.
• Enhance the performance of ROCm and HIP programs through techniques like coalescing, caching, prefetching, and profiling.

Course Structure

• Interactive lectures and discussions.
• Extensive exercises and practice sessions.
• Hands-on implementation within a live-lab environment.

Customization Options

• For requests regarding customized training for this course, please reach out to us to make arrangements.