What is time series analysis?
With our data now ingested into ADX, we can write queries using KQL to filter and search for data using search predicates. These queries are sufficient when your data is clean, structured, and you know what you are looking for, but what if we would like to do a more sophisticated exploration of our data? How can we look at trends, detect anomalies and outliers, and make forecasts, which, in turn, can help us make better decisions? One solution is time series analysis.
Note
The primary goal of this chapter is to focus on the practical aspects of time series analysis using the features provided by KQL. Although we will introduce the basics of time series analysis, an in-depth discussion is outside the scope of this book. There is a lot of good literature regarding time series analysis at https://www.packtpub.com.
Let's begin by discussing what a time series is. A time series is a sequence of data points captured over time. CPU, RAM, and disk...