Lindsey Heagy, a Postdoctoral researcher in the department of statistics at the University of California Berkeley, joins Matt Rocklin and Hugo Bowne-Anderson to discuss scientific computing in the geosciences with Python and Jupyter. Her research uses geophysical data to develop models of the subsurface for locating groundwater, characterizing mineral deposits, and environmental applications.
Research often starts with exploratory computing, where Jupyter serves us well! As we scale up our work to include larger data sets and more costly computations, we might move our work from a laptop to the cloud or a High Performance Computing (HPC) facility to access larger resources. And, as more research computing now starts with exploratory analysis of large datasets, we need tools that facilitate fluid interactivity. The combination of Jupyter and Dask are making it easier to achieve both of these tasks: moving from early algorithm development to larger-scale computations on different infrastructure, and providing avenues to visualize and explore large datasets on the cloud or HPC centers.
After attending, you’ll know
- How open tools in the Python ecosystem are used for geoscience research,
- Examples of using Dask on High Performance Computing systems (e.g. NERSC, Cheyenne),
- The Jupyter meets the Earth project and examples of how we are using and contributing to tools in the Pangeo and Jupyter ecosystems.
Join us this Thursday, September 24th at 5pm US Eastern time by signing up here to dive into the wonderful world of scientific computing in the geosciences with Python and Jupyter!