Come and Join the Chameleon Team!

We are hiring! Come and join the team to help us build and operate the kick-ass Chameleon experimental platform as well as develop new ideas in edge computing and reproducibility. You will be working on the bleeding edge, state-of-the-art technologies, evaluating and extending open source systems and tools, and developing new features -- and most importantly working with our amazing community of users who take our breath away every day with their cool resaerch ideas! 

 

Chameleon Changelog for November 2022

We hope everybody's had a great Thanksgiving! We've been busy bees this month supporting IndySCC and the many supercomputing projects and demos but still have a little something we can all be thankful for: we now have InfiniBand on the A100 nodes so you can build sophisticated data systems of components connected via RDMA fabric -- and we also have better filters for hardware so you can find those components more easily! 

Haswells are Back: an Interview with the UIC Associate Site

Today's Tips&Tricks blog spotlights one of the Chameleon Associate Sites: the Electronic Visualization Laboratory (EVL) at the University of Illinois Chicago. Learn about the hardware availability at the site, the motivation behind creating it, and some hands on insights gained in the process of deploying CHI in a Box on the site!

Zeus: GPU Energy as a First-Class Resource in DNN Training

In this month's user experiment blog we get a fascinating insight into how much power training deep neural networks (DNNs) consumes – and how to make it less. The authors’ discuss research presented as part of their NSDI ’23 paper, describe how they structured their experiments on Chameleon, and explain why bare metal resources are essential for power management research. 

Chameleon Changelog for October 2022

Things were pretty scary yesterday but we managed to pull a few tricks and bring you some new treats to enjoy in November! Between new Fugaku nodes, the ability to experiment with SGX, and a better way to work with networking at CHI@Edge, we hope to keep you busy and entertained this next month!
 

Fugaku Nodes now at CHI@TACC

Interested in the Fugaku Supercomputer? We now have 8 Fugaku nodes (Fujitsu FX700), available in CHI@TACC! Each of these nodes has a 48 core ARM A64FX CPU, 32 GiB of HBM2 memory, 512GB of NVMe storage, and HDR100 Infiniband. Notably, the high-bandwidth memory and non-x86 architecture are hard to find in other systems. TACC’s Frontera Supercomputer originally procured these for evaluation, but they’re now available for general use in Chameleon.