Method and system for real-time identification of blast radius of a fault in a globally distributed virtual desktop fabric

Abstract

A system and method for determining a blast radius of a major incident occurring in a virtual desktop system is disclosed. The virtual desktop system has interconnected service components and provides access to virtual desktops by client devices. An event collection module collects events from the service components. An aggregation module merges the collected events in a time-ordered stream, provides context to the events in the time-ordered stream through relationships between the collected events, and generates a correlated event stream. An analysis module determines a stream of problem reports from the correlated event stream. The analysis module determines a spike in the stream of problem reports and determines the attributes of the problem reports in the spike to define the major incident. The analysis module determines a scope of the major incident and a corresponding attribute, to determine a blast radius associated with the major incident in the desktop system.