Federal Project Grant Award Summary

The National Science Foundation's Division of Information and Intelligent Systems awarded a $358,200 Computer and Information Science and Engineering (CFDA 47.070) Project Grant to the University of Maryland, College Park, effective June 1, 2026 through May 31, 2031. This CAREER award funds the development of an approximation-first telemetry architecture for hyperscale networked systems that addresses the critical challenge of monitoring large-scale infrastructure supporting cloud computing, artificial intelligence services, and internet operations. The project delivers four integrated research thrusts: mergeable data summaries for end hosts and networked devices that track uncertainty; low-latency aggregation and query methods operating directly on compressed summaries; learning-guided compression techniques for long-term telemetry storage using both lossy and lossless approaches; and a management engine that optimizes telemetry configurations based on user requirements for accuracy, responsiveness, and operational cost.

The deliverables under this award advance monitoring capabilities for detecting network anomalies, bottlenecks, failures, and attacks while substantially reducing computational, storage, and energy overhead associated with traditional full-data telemetry approaches. The project also produces educational materials and hands-on learning opportunities in networking, cloud computing, and systems, with components planned for open-source release. These outputs serve the broader networking, cloud computing, and distributed systems communities by enabling more efficient and responsive monitoring of hyperscale networked infrastructure critical to national communication, commerce, education, health, and scientific operations.