A marker of success for any website is its ability to respond to user requests in an appropriate amount of time, when response times fall outside of what is normal it can lead to a negative experience for the end user or consumer. Increases in website response times could be attributed to any number of factors, or just simply a natural effect of demand during peak business hours. Being able to detect response time outliers in a series of what may seem like normal response time events can help website operators get ahead of potential issues before they become a greater problem.
In this recipe, you will write a Splunk search with the assistance of the Splunk Machine Learning Toolkit to detect outliers in server response times over a given period of time. The results will be visualized and added to a dashboard.
