Besides performance (unavailability of application as cited by crew to the helpdesk), revenue pilferage was one of the biggest pain point for the airlines to seek APM initiative. The crew would often site the unresponsiveness of the application, resulting in missed sales / revenue inputs from crew members after the flight lands. Because of these missed entries, the revenue won’t get recognized and the supply chain forecast for the eatables and other items went for a toss. Majority of times the issues were not being reproduced in the test environment. Also, there was very little visibility into the system as well as the network available at the airport, since it was outside the airline’s network and hence very little troubleshooting could happen.
Application Monitoring: Real time production monitoring to generate alerts upon unfavorable events, identify the bottlenecks with concurrent load, capture the top contributing queries, methods, etc. User Experience Monitoring: Identify exact user struggle, which can be reproduced within the test environment and fixed, since most of it was not even detected at the backend. End-to-end transaction mapping and correlation of complete data from client-side (browser) to application / server side, to logs. Idea was to have a consolidation and representation of performance metrics and correlated data in a single unified dashboard
providing ability to drill-down to root cause at any level.
Real-time alerts for unfavorable system stats (response time, CPU, memory, network, etc.) as well as for unfavorable user experience (form errors, page errors, high page load, etc.) to notify users for performance impacts. 24×7 End-to-end monitoring of crew portal resulted in identifying even those sessions where crew didn’t have any issues, it’s just that they were tired and wanted to avoid time filling the sales report, so as to reach their hotel early. However, there were certain performance issues arising out to DB queries. Analytics including top business transactions, flow path, exception / errors, method timing, etc. led to the discovery of issues. Correlation between very slow and error category transactions enabled with DB monitoring helped identifying the offending db queries. Cavisson helped airlines to improve the efficiency of crew portal by optimizing performance of the application.
As a leading organization, this airline needs to ensure maximum availability and performance of its mission-critical applications. Numerous applications in production environments have already been identified already and Cavisson is helping them to roll out monitoring in a phase-wise approach.
