Chameleon Office Hours for December 13, 2024

To provide support to our users, we are piloting by appointment Chameleon Office Hours. The first session will be on December 13, 2024, from 10:00-11:00 am and 2:30-3:30 pm Central time. Office hours appointments can be scheduled for 20 minute intervals, at least 8 hours in advance. To schedule an appointment, see this link.

As always, if you need help outside of this time, please contact our help desk. Additionally, you can find useful information about Chameleon on our blog, in our documentation, and on YouTube.

Chameleon Changelog for November 2024

This month we have presentations from the 5th Chameleon User Meeting on practical reproducibility, and lots of housekeeping on Chameleon. These include improvements to IP availability, and fixes to the core testbed, Trovi, and CHI@Edge.

Empowering the Edge: Breaking Heterogeneity Barriers in Cloud-based ML Training

Optimizing Federated Learning for Heterogeneous Edge Devices

Learn how researcher Redwan Khan uses Chameleon to develop FedCaSe, an innovative framework that tackles the challenges of distributed machine learning across diverse edge devices. This groundbreaking research demonstrates up to 29x improvement in client participation and 81x better data access efficiency, paving the way for more accessible and efficient AI systems.

Chameleon Takes Flight at SC24: Advancing Research and Collaboration

Read on to see what we'll be up to during SC24 week! Nov. 17 to 20

Join Chameleon at SC24 in Atlanta from November 17-20, 2024, as we showcase our contributions to advancing high-performance computing research. From hands-on tutorials to a workshop on reproducibility, BoF sessions, and REU student presentations, we invite you to explore Chameleon’s role in driving innovation. Don’t miss our integration with SCinet and FABRIC, opening new collaboration opportunities. For more details or to connect with our team, visit us at SC24!

Building MPI Clusters on Chameleon: A Practical Guide

Simplifying Distributed Computing Setup with Jupyter, Ansible, and Python

Running distributed applications across multiple nodes is a common need in scientific computing, but setting up MPI clusters can be challenging, especially in cloud environments. In this post, we'll explore a template for creating MPI clusters on Chameleon that handles the key configuration steps automatically, letting you focus on your research rather than infrastructure setup.

Full Agenda Out Now – Community Workshop on Reproducibility in ATL Nov. 18

We’re excited to announce the full agenda for the upcoming Community Workshop on Practical Reproducibility in HPC, taking place on November 18, 2024, at Terminus 330 in Atlanta, GA. Visit the workshop website for complete details.

Final Call for Early Bird Registration - Community Workshop on Practical Reproducibility in HPC (Nov. 18, 2024)

Where Theory Meets Practice: Transforming HPC Reproducibility from Aspiration to Action

Join us for a groundbreaking workshop that's shaping the future of HPC research reproducibility - and save $80 while you're at it! Early bird registration now extended until October 25th. Register Now.

If At First You Don't Succeed, Try, Try, Again...? Insights and LLM-Informed Tooling for Detecting Retry Bugs in Software Systems

Using Chameleon to Hunt Down Elusive Retry Bugs in Software Systems

Discover how Bogdan Stoica and researchers at the University of Chicago developed Wasabi, an innovative tool that combines fault injection, static analysis, and large language models to detect and analyze retry-related bugs in complex software systems. Learn how Chameleon's bare-metal capabilities enabled precise testing environments for this fascinating research published at SOSP'24.

Packaging Your Experiments on Chameleon with Python-chi 1.0

Streamlining Chameleon Experiments with Enhanced Python Library

Discover the powerful updates in Python-chi 1.0, Chameleon's official Python library. This new release brings improved typing, new modules for hardware and storage management, interactive Jupyter widgets, clearer error messages, and easier low-level access. Learn how these enhancements can simplify your experiment workflow, from resource allocation to data analysis. Whether you're a seasoned Chameleon user or just getting started, Python-chi 1.0 offers tools to make your research more efficient and reproducible.