Interactive 3d digital twin helps visualize supercomputer hardware and activity
Object Model Analysis of a Supercomputer with Digital Twin
Distributed, Parallel, and Cluster Computing
Summary
Supercomputers are complex machines made of many parts, and it can be hard to understand how all these parts work together or spot problems. The authors created a 3D digital twin, a virtual model, that shows the supercomputer’s physical setup like racks and network links, along with live activity and health information. This model runs in a game engine and lets users explore and select parts either by clicking in 3D or using commands, showing detailed stats and ongoing metrics. This helps operators and developers quickly connect performance or problems to the right hardware piece.
What this means in practice
- •For supercomputer operators: Locate and monitor hardware components visually within a 3D model to quickly diagnose performance and health issues in real time.
- •For data center managers: Use the digital twin to understand hardware layout and telemetry in one integrated interface improving maintenance and resource allocation.
Authors
Shilpika Shilpika, George K. Thiruvathukal, Venkatram Vishwanath, Michael E. Papka
Abstract
Operators and developers need a mental model of both the structure and the live behavior of a large supercomputer, but its physical layout, logical organization, and streams of per-node telemetry are difficult to relate to one another, making it hard to trace a metric or event back to a specific hardware component. We present DAT, an interactive three-dimensional digital analytics twin of a compute cluster built in a real-time game engine, Unreal Engine. DAT expands a compact, parametric description of a supercomputer, a reusable Digital Twin Prototype (DTP), into a navigable Digital Twin Instance (DTI) that mirrors its physical containment hierarchy of racks, chassis, blades, and network links, encoding each node's role and health in its appearance, while a lightweight event-driven simulator animates job and hardware activity over a virtual clock. Our current implementation adds a two-path node-selection mechanism, unifying direct 3D pointing with command-shell queries, that opens an in-world visual-analytics panel beside any selected component showing summary statistics and live, time-varying metrics. We describe this architecture, report qualitative behavior from the working prototype, and outline the path toward driving the panels with recorded telemetry and in-situ anomaly detection.