Real-time Analytics
Argonne and its collaborators needed to coordinate gravitational-wave and radio follow-up across observatories without moving proprietary datasets, but CPU-bound processing, siloed data-sharing policies, and slow cross-site workflows limited timely multi-messenger discovery.
Reported impact
RADAR enabled privacy-preserving observatory coordination, processed 4,096 seconds of advanced LIGO data in under 4.5 minutes, and achieved a 5-10x speedup versus CPU workloads so researchers can respond faster to multi-messenger events without breaking data-sharing rules.
Argonne and partner institutions built RADAR as an event-driven, federated framework for multi-messenger astrophysics and ran it on NVIDIA-accelerated supercomputing systems. Site-local AI inference processes advanced LIGO data and matches detections to public LIGO-Virgo-KAGRA superevents before Octopus publishes a trigger across the workflow, often using ProxyStore to move large intermediate results securely. RADAR then uses an AI parser to convert unstructured GCN circulars into structured metadata and combines those public alerts with private observatory measurements while keeping raw data local. Downstream, Dingo-BNS performs gravitational-wave parameter estimation and afterglowpy models federated radio afterglows. NVIDIA A100, A40, and GH200 GPU systems on Polaris, Delta, and DeltaAI supply the compute layer that makes this privacy-preserving inference and coordination pipeline fast enough for real-time follow-up.
RADAR performs site-local AI inference on LIGO data, matches detections with public superevents, sends triggers through Octopus and ProxyStore, parses GCN circulars into structured metadata, combines those alerts with local observatory measurements, and runs Dingo-BNS plus afterglowpy to refine source parameters without moving raw proprietary data.
Integrating NVIDIA GPUs on the Polaris, Delta, and DeltaAI supercomputers, the framework's AI module for signal detection processed over an hour—4,096 seconds—of advanced LIGO data in under 4.5 minutes.
Researchers at Argonne National Laboratory and collaborating institutions have developed RADAR, a federated, privacy-enhancing framework that enables observatories to coordinate gravitational-wave and radio follow-up without moving or exposing proprietary data.
Subsequently, RADAR integrates information across messengers by running Dingo-BNS for gravitational-wave parameter estimation and afterglowpy for federated radio afterglow modeling.
This translates to a speedup of 5-10x compared to the initial timeframe when running on CPU workloads.
When a matching superevent is found, RADAR publishes a trigger to Octopus—a cloud-hosted, event-driven messaging fabric—often using ProxyStore to handle large intermediate results efficiently and securely.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Oct 9, 2026 California Department of Transportation (Caltrans) Caltrans needed to inspect and monitor roughly 1,200 active transportation projects more frequently and safely, but manual site visits were slow, inconsistent, and risky for staff working around traffic, bridges, and… Remote operations up to 100 miles Esri ArcGIS · NVIDIA CUDA · NVIDIA Jetson Orin NVIDIA Sample Jun 29, 2026 Kaohsiung City Government Kaohsiung needed citywide situational awareness across transportation, flooding, disaster response, and public-safety workflows, but departmental silos and traditional computer vision left agencies unable to interpret… 30,000 diverse smart city camera streams integrated NVIDIA Cosmos · NVIDIA DGX · NVIDIA Metropolis Blueprint for Video Search and Summarization NVIDIA Sample Mar 4, 2026 Israel Meteorological Service The Israel Meteorological Service needed to produce faster and more localized weather forecasts for floods, fire seasons, and other severe conditions, but conventional numerical weather prediction required large CPU… 2.5-kilometer resolution forecasts NVIDIA Earth-2 · NVIDIA Earth-2 Corrector Diffusion (CorrDiff) · NVIDIA PhysicsNeMo 