Streaming real-time data from Telegraf to Grafana

In previous chapters, we typically sourced a collection of metrics from, say, a website or a data file, and after storing that data in a time-series database, we then used Grafana to query that data for analysis and visualization. This process of working a corpus of data as a single entity is sometimes referred to as batch handling. By contrast, in this chapter, we will be continuously moving time samples of data as they are generated, or streaming the data from source to destination.

This style of handling data is suitable for many applications where the data needs to be visualized or otherwise processed in real time for monitoring and alerting. It is also most useful in those cases where it would be difficult or impractical to wait for all the data to arrive before analyzing, or where sampled, non-transactional data is the norm.

To get an idea about how to distinguish between these two use cases, imagine the case where you need...