Aparapi: An Open Source tool for extending the Java promise of ‘Write Once Run Anywhere’ to include the GPU

Aparapi provides an API for expressing data parallel workloads in Java and a runtime capable of converting Java bytecode into OpenCL for execution on a GPU. For some data-parallel algorithms, executing on the GPU can offer substantial performance gains.

We will introduce Aparapi, demonstrate key features as well as discuss lessons learned during the transition from closed to open-source.