The three pillars of observability, namely logs, metrics, and traces, have been raising our industry’s awareness of the three categorical data sources that are indispensable to implement true observability. The need for observability becomes even more obvious and pressing as an organization’s infrastructure and workloads evolve to be more hybrid, distributed, and, inevitably, complex. However, in reality, DevOps professionals a lot of times still face the challenges diversified tool chains and disparate data silos. It is no secret that every single signal from applications or infrastructure in itself is limited in its scope and usefulness. The true power only comes in when logs, metrics, and traces are properly combined and correlated together.
Everyone is talking about it, but only a handful few really knows the devil details. In this talk, we will show various scenarios where the combination of logs, metrics, and traces truly make a difference. We will also review some of the challenges and best practices of integrating the three pillars together using OSS tools (and its ecosystem) such as the Elastic Stack.