Cloud Native Monitoring

Conclusion

The monitoring and observability landscape has changed greatly over the past three to five years. System architectures are sufficiently different from their pre–cloud native counterparts to demand a new paradigm, born from radically rethinking, as an industry, how we build and implement monitoring systems.

This kind of radical rethinking allows new ideas to emerge. One such idea, as you’ve learned, is Sridharan’s “three pillars of observability.” Although it’s a flawed paradigm that focuses on the data rather than the outcome, it’s also an inspiring glimpse into what’s possible.

To refine our focus and make a discernible impact, our thinking about cloud native monitoring must pivot away from the “three pillars” and toward the “three phases of observability” we’ve outlined in this report. These three phases allow for a goal-driven, pragmatic approach to cloud native monitoring that emphasizes remediating problems and improving business outcomes.

Open source solutions continue to drive innovation, even as more enterprises adopt cloud native monitoring. The companies encouraging them to do so, like Chronosphere, Grafana Labs, Weaveworks, Amazon, and Google, are adopting more and more open source technologies, as well as creating and using frameworks that allow cross-compatibility between various technologies and organizations.

We’ve emphasized metrics in this report because they can provide a high-level overview of everything happening in your system, at a fraction of ...