Intro to HPC Bootcamp Projects

This project introduces fusion energy as a sustainable solution to growing global energy demands and climate change. Participants will study the basics of fusion, including plasma physics and modeling, and use simulations to explore reactor efficiency and address the challenges of practical energy production.

This project focuses on understanding and optimizing energy efficiency in High-Performance Computing (HPC) facilities, addressing the significant energy demands of modern computing systems. Participants will gain practical skills in monitoring energy use, analyzing power consumption patterns, and applying optimization techniques to enhance sustainability and cost-effectiveness in real-world HPC environments.

The DOE's EAGLE-I system, managed by Oak Ridge National Laboratory, provides real-time data on energy infrastructure and disruptions to support emergency response. This project aims to develop tools to analyze the characteristics and causes of power outages, helping improve disaster response and recovery efforts.

This project explores short-term load forecasting (STLF) using deep learning models to predict electricity demand in buildings, leveraging the BuildingsBench platform developed by the National Renewable Energy Laboratory. 

This project will explore using large language models for scientific applications. We will train entirely from scratch a (small) language model that is able to generate coherent text. Following this, we will explore different scaling behaviors of large language models and explore how they can be applied to scientific problems.

This project engages undergraduate students in learning high-performance computing (HPC) by using large language models (LLMs) to generate and evaluate instructional content. Students will build foundational HPC knowledge while critically analyzing the accuracy and effectiveness of AI-generated explanations compared to expert sources.

This project explores organic dye-sensitized solar cells (ODSSCs) as a low-cost and flexible clean energy technology. Students will use data science and machine learning to analyze molecular datasets and predict properties of eco-friendly materials for ODSSCs, with a focus on affordable energy production and material discovery.