Arkouda: NumPy-Like Arrays at Massive Scale Backed by Chapel

Workshop: Arkouda: NumPy-Like Arrays at Massive Scale Backed by Chapel

Abstract: Exploratory data analysis (EDA) is a prerequisite for all data science, as illustrated by the ubiquity of Jupyter notebooks, the preferred interface for EDA among data scientists. In order to expand the scale of EDA, we have designed and built a software package for using NumPy-like arrays and Pandas-like data frames at massive scale. We have also integrated this software package into our data science workflow. This software package, called Arkouda, is primarily implemented in the Chapel programming language. Arkouda is currently in process to be open-sourced so our work can be shared with a broader community. In the proposed talk we will cover the why, how, and what of Arkouda.

Back to Parallel Applications Workshop, Alternatives to MPI+X Archive Listing

Back to Full Workshop Archive Listing