Research Timeline

Learned the fundamentals of Edge computing and reviewed related research to establish context.

Defined a roadmap for creating Edge benchmarks and learned to use Orbit for node management.

Set up monitoring and development tools, created a reusable environment image, and explored running LLMs on GPUs.

Built a pipeline to collect and store live metrics, designed a standardized LLM test, and launched a project website.

Refined the roadmap, set up exporter nodes, measured their overhead, and began initial system benchmarking.

Completed GPU stress testing, developed a live metrics collection script, and established baseline GPU benchmarks.

Implemented an asynchronous pipeline for testing LLMs with GPU metrics, ran sample tests, and performed exploratory data analysis.

Analyzing collected data and developing a benchmark process to created a weighted-sum number to represent computer power.

Compiling research findings and preparing comprehensive documentation.

Presenting research findings and recommendations to stakeholders.