Google Summer of Code

For the first time, Forschungszentrum Jülich will be participating in Google Summer of Code as a mentoring organization with the JSC-led project Heat. Initially funded within the Helmholtz Analytics Framework (a Helmholtz Information & Data Science pilot project), Heat is an open-source Python library developed with the goal of enabling data-intensive research – multi-CPU and multi-GPU computing, and highly optimized machine learning and deep learning – using a NumPy/SciPy-compatible API. Its general, easy-to-adopt solutions to single-node memory bottlenecks have led to an expansion of interest and feature requests from outside the Helmholtz Association.

Google Summer of Code (GSoC) is a worldwide online programme that aims to introduce new contributors to open-source software development. Over the course of the summer, GSoC contributors mentored by members of the Heat core team (from JSC, DLR, and KIT) will work on implementing new features to support our new and established user communities, from memory-distributed signal processing to distributed sparse operations, parallel dimensionality reduction algorithms, expanded support of hardware accelerators, and more.

Interested participants are already contributing to Heat's repository, engaging on our Mattermost channel, and drafting proposals. Coding officially starts on 14 June, preceded by a three-week Community Bonding Period. We relish the opportunity to work with open-source enthusiasts, support our expanding user base, and act as pathfinders for future GSoC participation by other open-source projects under the FZJ umbrella. More information about Heat can be found on GitHub: https://github.com/helmholtz-analytics/heat.

Contact: Dr. Claudia Comito

from JSC News No. 288, 29 April 2022