Robot Based Thermal Imaging Monitoring and Omniverse Dashboard Development in Data Centers
Tuesday, June 10, 2025 3:00 PM to Thursday, June 12, 2025 4:00 PM · 2 days 1 hr. (Europe/Berlin)
Foyer D-G - 2nd floor
Project Poster
Data Center Infrastructure and CoolingDigital Twins and MLIndustrial Use Cases of HPC, ML and QCSustainability and Energy EfficiencyVisualization and Virtual Reality
Information
Poster is on display.
Efficient temperature management in data centers is critical as server density increases. Traditional monitoring systems rely on fixed sensors, which are limited in detecting anomalies and providing a comprehensive view of thermal distribution. This research presents a robot-based thermal imaging system for real-time temperature monitoring in data centers. Using probabilistic methods, thermal data is accurately mapped to server components, even in the presence of positional errors. Furthermore, the integration of NVIDIA Omniverse enables advanced 3D visualization, providing operators with an intuitive and detailed view of temperature distributions. This approach not only enhances decision-making and resource optimization but also lays the foundation for future integration with AI-driven prediction models and scheduling systems. By enabling predictive and automated temperature management, the proposed system offers a scalable solution to improve reliability and efficiency in data center operations.
Contributors:
Efficient temperature management in data centers is critical as server density increases. Traditional monitoring systems rely on fixed sensors, which are limited in detecting anomalies and providing a comprehensive view of thermal distribution. This research presents a robot-based thermal imaging system for real-time temperature monitoring in data centers. Using probabilistic methods, thermal data is accurately mapped to server components, even in the presence of positional errors. Furthermore, the integration of NVIDIA Omniverse enables advanced 3D visualization, providing operators with an intuitive and detailed view of temperature distributions. This approach not only enhances decision-making and resource optimization but also lays the foundation for future integration with AI-driven prediction models and scheduling systems. By enabling predictive and automated temperature management, the proposed system offers a scalable solution to improve reliability and efficiency in data center operations.
Contributors:
Format
On DemandOn Site
