From the course: Microservices: Design Patterns (2020)

Unlock the full course today

Join today to access over 22,400 courses taught by industry experts or purchase this course individually.

Metrics aggregation patterns

Metrics aggregation patterns

From the course: Microservices: Design Patterns (2020)

Start my 1-month free trial

Metrics aggregation patterns

- [Instructor] In a similar vein to log aggregation, operationally, metrics are invaluable to diagnosing issues. In fact, metrics can be more powerful than logging if used correctly. Metrics is also easier than logging because there's really less human interaction. Logging relies on a developer to write and structure the log message. Metrics, however, usually just requires a bit of instrumentation. The problem we're looking to solve is that we need to know what is going on with the system or service at a system level. We aren't really looking at code output, we're looking for system output. Once again, a common taxonomy is critical with metrics. Structuring your keys to be consistent and descriptive will help explain what the ultimate dashboards are saying. As with logging, much has been done already, and there are standard metrics libraries available in almost every language to generate metrics. While much of the metrics…

Contents