It's our post-SC wrapup! In this month's changelog, we've got several new edge devices to announce, as well as improvements for UEFI boot support, a better way to view and track your allocation spending, and more.
Increasingly more computer science systems conferences are using Chameleon for artifact evaluation. Learn more about how to use Chameleon for reproducibility; including tips both for organizers of these events and artifact authors.
Researchers at the University of Missouri explore novel middle-man data poisoning attacks with facial authentication applications and propose a defense architecture. Learn more about these attacks, their solution, and the researchers themselves in November's featured user experiments blog.
This year Chameleon hosted the IndySCC competition, analogous to the in-person Student Cluster Competition (SCC) of Supercomputing. Teams use cloud/shared resources to optimize a variety of HPC workloads in order to complete the most computations during a 48 hour final, all while staying below a strict power cap. Learn more about the compettion, Chameleon's support, and where you can see the results.
Attending SC21? Check out the initiatives Chameleon is supporting including workshops, papers written using Chameleon, and more! If you’re presenting something at SC21 and not on this list, send us an email!
In this month's changelog, we're excited to announce a new batch of hardware at CHI@UC, including more flexible network topologies, some very high-powered nodes, and InfiniBand-connected GPUs. We've also added support for provisioning ARM-based machines to support CHI-in-a-Box associate sites who wish to configure ARM nodes. As always, more inside!
The recently released Chameleon Phase 3 hardware will support new experiments in networking, disaggregated hardware, and more. Learn more about the different types of hardware and what kind of experiments they're best suited for!
Learn about University of Chicago PhD student Chengcheng Wan's research on designing a solution to integrate machine learning components into traditional software systems to maximize energy efficiency, latency, and accuracy constraints at machine learning, system and application levels, with related research published at USENIX ATC, ICML, and ICSE.
In this month's changelog, we announce a few really cool new features, one of which has been on our most-requested list for a long while: PI delegate! Besides that, there is a brand-new preview capability that allows you to share your Trovi experiments with non-Chameleon users so they can reproduce your work w/o an allocation. More details, as always, inside.
Join us on Friday, October 22 at 1.30p CT for Chameleon Live - an informal webinar where the Chameleon team will showcase the newest features of the testbed and answer questions from the community. Think of it as office hours to learn all about Chameleon updates and how to use them! Register and submit questions here: https://forms.gle/4itPs913vFTbnnGG9