Profiling is an effective way of understanding which parts of your application are consuming the most resources. Traditionally, logs, metrics and traces have been considered the three pillars of observability, but more recently profiling has emerged as a fourth pillar to be used alongside these other observability tools.

Continuous Profiling, in particular, adds a dimension of time that allows you to understand your system’s resource usage (i.e. CPU, Memory, etc.) over time and gives you the ability to locate, debug, and fix issues related to performance.

In this talk, we'll present Pyroscope, an Open Source Continuous Profiling platform, explain the particular challenges it needs to tackle and showcase how it can be used to analyze and fix performance bottlenecks:

• Describe what Continuous Profiling is and what it brings to the observability universe.
• Present Pyroscope, an Open Source Continuous Profiling platform, on a high level.
• Dig deeper into Pyroscope's internals, explaining the particularities of its storage engine and how it allows doing Continuous Profiling at scale.
• Showcase specific use cases and examples where profiling can be and has been used to detect, analyze and fix performance bottlenecks.