ABSTRACT. Collaboration between domain scientists and software engineers produces novel software tools that enable researchers to investigate research questions, drive community engagement, and publish results with a shorter turnaround time, and at a larger scale than ever before. Two projects related to Green Energy at ORNL are presented to illustrate a real-world example of this type of successful collaboration. One is a Fusion simulator called GITR, used by researchers to examine how plasma-facing components in prototype Fusion reactor designs deteriorate under the harsh conditions. The other is a simulator to examine the behavior of green electrical grid component designs such as solar power plants in extremely large scale power networks.

ABSTRACT. In the realm of scientific software development, adherence to best practices is often advocated. However, implementing these can be challenging due to differing opinions. Certain aspects, such as software licenses and naming conventions, are typically left to the discretion of the development team. Our team has established a set of preferred practices, informed by, but not limited to, widely accepted best practices. These preferred practices are derived from our understanding of the specific contexts and user needs we cater to. To facilitate the dissemination of these practices among our team and foster standardization with collaborating domain scientists, we have created a project template for Python projects. This template serves as a platform for discussing the implementation of various decisions. This paper will succinctly delineate the components that constitute an effective project template and elucidate the advantages of consolidating preferred practices in such a manner.

ABSTRACT. The purpose of Software Quality Assurance (SQA) is to help prevent potential negative impacts of software failure. These impacts could entail an injury/illness event, a radiological hazard, an environmental insult, non-compliance with a rule/regulation, a security/cyber-security breach, loss of time and/or money, a negative reputation against ORNL, or any combination of these impacts.. Therefore, all staff who procure, develop, modify, or maintain software at the Laboratory – whether developed in-house, licensed from a commercial vendor, obtained from another organization, or otherwise acquired - are required to comply with the SQA requirements found within the SBMS.

ABSTRACT. Our laboratory develops novel technologies for advanced mass spectrometry analysis. These technologies only exist at ORNL. Most of the time, the combination of instruments requires customized software packages to synchronize the instruments and to analyze the collected data. The talk will showcase some of the software packages that allow unprecedented high throughput chemical analysis of liquid samples and single cells.

ABSTRACT. In this talk, the significance of incorporation of uncertainty quantification (UQ) approaches towards development of machine learning interatomic potentials (MLIPs) will be emphasized. In doing so, various UQ metrics will be shown to reduce data need, time, and energy consumption associated with MLIPs development, while enhancing the model’s interpretability. A workflow aimed at accelerating the development of transferable, robust and trustworthy MLIP potentials for several ORNL-centric material systems will be proposed for ORNL Leadership Computing Facility (OLCF) facilities.

ABSTRACT. Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.

ABSTRACT. This hackathon will organize participants to work on utilizing NN potentials for molecular simulation. The details below have been shared with and drawn some interest from group leaders in both PSD/Chem. Sciences Division and CCSD/CSED.

---

The goal is to build a knowledge base of methods for practitioners to get quickly to the point where they can run molecular dynamics simulations for a new liquid, solid, polymer or small molecule. No specific framework is preferred, but problems should be ORNL and DOE-relevant.

4 Sub-goals:

1. Fast QM calculation of forces and energies 2. Loading and running an ML potential for computing forces and energies 3. Data fitting from QM to ML. 4. Running MD with an ML potential

For each, the hackathon can be considered "complete" when a walk-through exists showing how to accomplish that step on an example system that has actually worked in our hands. As facilitator for the hackathon, I'll prepare links to similar "how-to" materials and some starting structures.

ABSTRACT. This tutorial will mostly be focused on giving an overview of the basics of designing and coding a Trame application that could be integrated into the NDIP platform. Trame is a web application framework that allows developers to use Python to develop Vuetify applications. Topics will include setting up a Trame project, an overview of the API, design patterns, integrating Trame with Galaxy, UI tips and tricks, adding authentication, and other related topics. The idea is to just get people started with Trame who have never heard of it, and more complicated subjects like visualization will not be covered in this presentation. The tutorial will be about an hour in length, and a sample project will be provided for attendees to use and to follow along with. I (Greg Cage) will be presenting.

ABSTRACT. The 2-hour hands-on tutorial will cover fundamental aspects of the Julia programming language and the JACC.jl package developed at Oak Ridge National Laboratory for performance portable CPU/GPU code. Built on LLVM, Julia is a language targeting scientific use-cases and has a rich ecosystem for GPU programming. The audience will be exposed to writing "parallel_for" and "parallel_reduce" patterns with actual code they can run and test on their systems.