Leo Han, a second-year Ph.D. student at Cornell Tech, conducted pioneering research on the fair attribution of cloud carbon emissions, resulting in the development of Fair-CO2. Enabled by the unique bare-metal capabilities and flexible environment of Chameleon Cloud, this work tackles the critical issue of accurately distributing responsibility for carbon emissions in cloud computing. This research underscores the potential of adaptable testbeds like Chameleon in advancing sustainability in technology.

High-resolution Vision-Language Models (VLMs) offer impressive accuracy but come with significant computational costs—processing thousands of tokens per image can consume 5GB of GPU memory and add 15 seconds of latency. The HiRED (High-Resolution Early Dropping) framework addresses this challenge by intelligently selecting only the most informative visual tokens based on attention patterns. By keeping just 20% of tokens, researchers achieved a 4.7× throughput increase and 78% latency reduction while maintaining accuracy across vision tasks. This research, conducted on Chameleon's infrastructure using RTX 6000 and A100 GPUs, demonstrates how thoughtful optimization can make advanced AI more accessible and affordable.

In this interview, Weichen Li, a PhD student from the University of Chicago discusses research on improving code editing through explicit transformation rules. EditLord breaks down the code editing process into clear, step-by-step transformations, significantly enhancing editing performance, robustness, and functional correctness compared to existing methods.

Scientific workflows often fail in unexpected ways, but traditional detection systems require massive amounts of training data. This groundbreaking approach generates just the right data needed to train anomaly detection models, improving accuracy while reducing resource consumption.

The AutoAppendix project evaluates computational artifact reproducibility across SC24 conference submissions, revealing that most researchers struggle with creating truly replicable experiments despite their importance to scientific validity. By developing one-click reproduction templates for the Chameleon Cloud platform, this research aims to transform how computational scientists share and validate their work, potentially saving countless hours of frustration for both authors and reviewers.

Processing large-scale genomics data efficiently is a monumental task, often hindered by high costs and resource allocation challenges. This blog dives into an innovative system designed to optimize genomics workflows by minimizing out-of-memory failures—a critical bottleneck in such operations. Through a combination of scalable benchmarking tools and a failure-aware scheduler, researchers are unlocking new possibilities for resource efficiency and reliability. Leveraging insights from Chameleon, this solution paves the way for groundbreaking advancements in genomic data processing.

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.

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.

Learn how cutting-edge research is shedding light on the energy dynamics of I/O operations in HPC environments, potentially reshaping future storage designs.

Martin Putra, a 4th year PhD student at the University of Chicago, shares how he used Chameleon to build and test a scalable benchmarking tool for genomics workflows, uncovering insights that could lead to more efficient resource management for these computationally intensive tasks.