Alpaka - An Abstraction Library for Parallel Kernel Acceleration

TL;DR

Alpaka is a C++ library that provides an abstract parallelism model to simplify porting applications across diverse hardware accelerators, enhancing portability and reducing development effort.

Contribution

It introduces a hierarchical parallelism abstraction that enables platform-independent programming and minimal code changes for new hardware support.

Findings

Supports various hardware types including CPUs and GPUs

Reduces code modifications needed for new platforms

Enables performance portability across accelerators

Abstract

Porting applications to new hardware or programming models is a tedious and error prone process. Every help that eases these burdens is saving developer time that can then be invested into the advancement of the application itself instead of preserving the status-quo on a new platform. The Alpaka library defines and implements an abstract hierarchical redundant parallelism model. The model exploits parallelism and memory hierarchies on a node at all levels available in current hardware. By doing so, it allows to achieve platform and performance portability across various types of accelerators by ignoring specific unsupported levels and utilizing only the ones supported on a specific accelerator. All hardware types (multi- and many-core CPUs, GPUs and other accelerators) are supported for and can be programmed in the same way. The Alpaka C++ template interface allows for straightforward extension of the library to support other accelerators and specialization of its internals for optimization. Running Alpaka applications on a new (and supported) platform requires the change of only one source code line instead of a lot of \#ifdefs.