Speaker
Description
We present a modular alarm and visualization framework designed to detect and interpret network anomalies that lead to performance degradation in WLCG infrastructures. The system consists of two interoperable components: Alarms And Alerts System, a Kubernetes-based backend that ingests perfSONAR measurements and automatically identifies routing changes, performance degradations, and related anomalies; and pSDash,a lightweight and extensible web interface for interactive visualization, correlation, and inspection of active alarms.
Together, these developments enhance the visibility, reliability, and diagnostic capacity of the WLCG network infrastructure. The platform has already been deployed in production environments and serves as a foundation for integrating advanced detection methods.