Welcome back to school Chameleonaires! As we say goodby to summer and get ready to head back to the confines of our offices, classrooms, and cubicles, take a few moments to check out some new and improved resources for getting started with Chameleon. Our engineers have also been busy this summer, cooking up new tutorials, including interactive Trovi artifacts digging into CHI@Edge’s recent releases (we’ve officially transitioned to production!) and live tutorials for your consumption.
Hello, Chameleon! Let’s Get Reacquainted
The Chameleon Testbed is a powerful tool for your research. But powerful tools also have complexity. That’s why we publish our Getting Started tutorial in our Chameleon documentation. The tutorial walks you through the complete process of getting an instance up and running on Chameleon hardware.
We’ve updated the guide to:
- Include a brand new introduction on using Python-Chi and the Chameleon Jupyter interface to reserve resources and launch instances programmatically.
- Improve instructions for logging into Chameleon for the first time and setting up a project.
- Clarify “Create Lease” and “Launch Instance” actions in the Chameleon GUI.
- Note common pitfalls to avoid when working with Chameleon resources.
Whether you are introducing someone new to Chameleon, or just want a refresher yourself, the revamped Getting Started guide is available to help move your research forward.
We are always looking to improve. If you have comments or suggestions for us, regarding our Getting Started guide, other documentation, or the testbed itself, please reach out at contact@chameleoncloud.org.
New Trovi Artifacts & Tutorials!
If you’ve missed our last few tips and tricks, we published some new updates on CHI@Edge that you should check out! These blogs include links to multiple artifacts and resources to help you unleash the power of CHI@Edge for your own projects. Highlights include a demonstration of how to SSH to a CHI@Edge node, methods for connecting peripherals (like cameras) to a CHI@Edge node, and how to run GPU workloads on CHI@Edge nodes.
CHI@Edge Blogs:
We also published some tips on how to use the new Layer 3 connection on FABRIC and how to set up power measurement and management on Chameleon.
Finally, we had a fun start to the summer with the return of our Summer on Chameleon series. Check out our updated and refreshed tutorials on Chameleon orchestration, reservations, and a deep dive on CHI@Edge on our Youtube channel.
Oldies, but Goodies
Don’t forget to check out our other popular resources and tutorials, as well as some of the interesting user-contributed content that you can find in the treasure trove of Trovi.
Useful Experiment Patterns
- Bare Metal CPU
- Bare Metal GPU
- MPI Cluster
- KVM Experiment Pattern
- Deploying a Kubernetes Cluster
- CHI-in-a-Box QuickStart
- chameleon_fabnetv4
Reproducible and Educational Experiments
- Reproducing “Deep Neural Nets: 33 years ago and 33 years from now”
- AutoLearn Autonomous Cars
- Data Acquisition - Locally Generated Data
- NanoGPT Toy Version
- Baleen: ML Admission & Prefetching for Flash Caches (FAST 2024)
- Metis: File System Model Checking via Versatile Input and State Exploration (FAST 2024)
We hope these resources will boost your efficiency and familiarity with the testbed. As always, if you need additional or personalized help, you can reach our engineers through the Help Desk.
Happy experimenting!