Yellow is the New Green: Meaningful Indicator Threshold from the Operations Perspective

Enterprise Application Monitoring is an essential task to support Operations. The downsides are the high effort of data collection as well as Application Architecture documentation. While hard failures are easy to identify – an application just stop working, it is a great challenge to identify reductions in service warranty fulfillment. Especially for applications with over-fulfillment of defined Service Level [Agreements] (SLA) it can be tight to identify noteworthy reductions of performance. As a result, an Operations team has the challenge to notice a constraint while the status of an application is “green” based on the SLA. From an Operations perspective it is essential to identify potential incidents as early as possible to initiate counteractions.This paper discuss based on a Case Study, how patterns can be identified to define meaningful values for the “yellow” threshold of an application in order to support Service Management and Operation. The goal is to utilize existing data to minimize the impact to the IT infrastructure. Using existing log data from applications and monitoring solutions help to achieve this goal by limiting the additional efforts to analyzing and reporting.