Theme: Evolution of software development and management
Keeping Things Consistent: Developing a Framework to Unify Future WMKO DRPs
While there are many wonderful instruments producing exciting and varied science, they each have their own ways of reducing data even though the underlying processes are pretty much the same. This leads to effort wasted on rewriting code that has already been written, inefficient processes using archaic infrastructure, and increased complexity in trying to reproduce refereed work. This poster will analyze the effort going into developing a generalized pipeline framework to facilitate the reduction of Keck data so instrument developers can have a unified approach to reduce data while allowing them to work into the pipeline their instrument’s unique curiosities. The modular approach will allow pipeline developers to reuse code for specific processes that are common among data reductions. It will talk about the design choices made during the design process and how choosing Python helped to focus the development effort. The Framework will utilize many astronomy tools in Python in order to reduce the amount of code required to be written. Finally, the poster will examine the approach being taken to record the process history for scientific reproducibility.
