DesignSafe Portal: A Smarter, User-Centered Dashboard for Natural Hazards Research - BYOP

Science Gateways Bring Your Own Portal (BYOP) Submission DesignSafe Dashboard Portal: h4ps://www.designsafe-ci.org/ Abstract: As part of the SGX3 Internship Program at the Texas Advanced Computing Center (TACC), our team collaborated with the DesignSafe group to reimagine and enhance their researcher dashboard. DesignSafe is a widely used science gateway supporting natural hazards research by providing access to HPC systems, data, and tools. The dashboard serves as the main entry point for researchers to manage jobs, allocations, and storage. Our work focused on modernizing the dashboard interface to make it more intuitive, informative, and helpful. Beyond design, we also introduced two intelligent backend features: one to predict the likelihood of job failure before submission, and another to assist users with cleaning and managing their data using machine learning. These improvements aim to reduce researcher frustration, increase productivity, and create a smoother interaction with the gateway. Key Features of the DesignSafe Modernized Portal: 1. Redesigned Dashboard Experience: The new dashboard interface provides a clean, modular layout showing recent jobs, allocations, favorite tools, and system status all in one place. Unlike the older version, which required users to dig into multiple sections, the updated design offers researchers an at-a-glance overview of everything they need to manage their work efficiently. 2. Job Failure Prediction using Machine Learning: To support researchers in reducing failed job submissions, we developed a machine learning-based feature prototype that analyzes job parameters and estimates the likelihood of failure before the job is submitted. Built using models like XGBoost and CatBoost, this feature parses historical job logs and provides a real-time risk score during the submission process. This directly improves productivity and saves valuable HPC resources. 3. Smart Storage Analyzer for Data Cleanup: We also implemented a storage analysis tool integrated into the Data Depot. The tool scans user directories, detects duplicate or outdated files, and provides cleanup suggestions based on machine learning (using Isolation Forest). The UI allows users to view flagged files and take cleanup actions with a few clicks. 4. Personalized Experience with Favorite Apps and Guides: The inclusion of user-specific favorites and embedded guides/tutorials makes the platform more personalized. Researchers can pin their most-used apps and access visual guides without needing external help or documentation. 5. Real-Time System Status and Tool Access: The dashboard now includes real-time system load indicators through Tapis API and quick access to recently used tools. This enhances responsiveness and helps researchers plan job submissions around system availability, improving turnaround times for their experiments. In conclusion, the redesigned DesignSafe portal brings together modern UI improvements and intelligent backend tools to support researchers in high-performance computing environments. With features like real-time system visibility, job failure prediction, and AI-assisted storage management, the portal offers a smarter, more efficient gateway experience. These upgrades help researchers focus on their scientific work while reducing overhead and system errors. This work demonstrates how thoughtful design and applied machine learning can enhance usability across large-scale research platforms. PowerPoint slide providing details of the gateway in visual (below): 7/30/25