Join GitHub today
Programming, Compiling and Deploying nBody Code with the POLCA toolbox
The POLCA tool box offers the full support to compile (annotated) C code to different target platforms, ranging from CPUs to FPGAs. With its annotation model, legacy code can be easily extended and readied for POLCA so that existing code becomes more portable and easier to maintain.
Since POLCA generates enriched source code, standard compilers can be used for the final conversion step. At the time of writing this, POCLA can generate OpenMP, OpenCL, MaxJ, FlexaWare and VHDL code. In the following we will demonstrate how POLCA supports development and compilation of a simple n-Body code:
Describes how to install the tools to set up a demo that you can run on your own machine.
In this step we will show you how to use the tools to write and annotate a simple n-body simulation.
POLCA offers you the right tools to assess the performance of a code without necessitating its compilation and deployment.
What are "transformations"? And what does POLCA do with the annotated code?
This step describes how you can generate code for CPUs, GPUs or FPGAs using the POLCA tools.
Testing the resulting code is easy with these POLCA steps.