Heterogeneous Computing Tools

Heterogeneous computing refers to systems that use more than one kind of processor. Heterogeneous systems architecture or HSA creates an improved processor design that exposes the benefits and capabilities of mainstream programmable compute elements. Explore AMD heterogeneous computing development tools, SDKs and solutions below or learn more about heterogeneous computing in our resources section.

Tools & SDKs

CodeXL — CodeXL is a comprehensive, open source tool suite that enables developers to harness the benefits of AMD CPUs, GPUs and APUs.

ROCm – The Radeon Open Compute platform delivers on the vision of the Boltzmann Initiative, bringing new opportunities in GPU Computing Research.

APP SDK — The APP SDK is a complete development platform to allow you to quickly and easily develop applications that are accelerated using OpenCL™. This SDK works with AMD heterogeneous computing technologies (AMD APUs or GPUs).

Legacy Heterogeneous Computing Tools — These tools have reached end of life and are no longer supported, but they, and the supporting documentation, are still available for download if you need them.


ACL — The AMD Compute Libraries are a set of open source solutions, providing developers with open source libraries targeted to those who want to  accelerate computations on GPUs, APUs and CPUs.

Aparapi — This API converts Java bytecode to OpenCL at runtime and executes it on the GPU. If Aparapi can’t execute on the GPU, it will execute in a Java thread pool.

Bolt — Bolt provides an STL compatible library of high level constructs for creating accelerated data parallel applications. Bolt includes an array of Bolt capabilities by including support for common compute-optimized routines including sort, scan, transform, and reduce operations. In its open-source debut, Bolt supports C++ AMP in addition to OpenCL™ as underlying supported compute technologies.

GPUPerfAPI — GPUPerfAPI is a library that can be integrated directly into your own graphics or compute application for accessing GPU performance counters. It requires AMD Radeon™ HDseries graphic cards.


Project Boltzmann HIP Datasheet – It’s HIP to be Open. Convert  your Cuda code to C++ using AMD’s new HIP tool. Learn more. Visit the GPUOpen website.

Everything  you Always Wanted to Know about HSA (But Were Afraid to Ask) – Whitepaper, by Nathan Brookwood, Research Fellow, Insight 64

HSA: A New Architecture for Heterogeneous Computing – Whitepaper/Research by Tirias Research, Sponsored by AMD

Visit the AMD OpenCL™ forum and talk with other developers working with OpenCL™