Brian Brazil

The OpenMetrics format intends to standardise metric exposition, making it easy for both those developing and operating systems to monitor them. It is however a new format. Will it be supported by your monitoring system? Will you need to rewrite your existing instrumentation? What's needed to transition? What about 3rd party systems you don't control? How does this differ and expand, and improve on the existing Prometheus format? This session will cover all of these questions.

Prometheus client libraries are notably different from most other options in the space. In order to get the best insights into your applications it helps to know how they are designed, and why they are designed that way. This talk will look at how client libraries are structured, how that makes them easy to use, some tips for instrumentation, and why you should use them even if you aren't using Prometheus.

As the industry moves towards more cloud based and containerised solutions such as Kubernetes, monitoring tools have to keep up. These new environments are far more dynamic than the hand-maintained machines of old, requiring more sophisticated and scalable approaches. This talk will look at how Prometheus has evolved over the past 5 years to be better able to cope with these challenges, including the 2.0 release and practices that we encourage in a cloud native world.

This talk will cover the dynamic nature of cloud-based services, and how approaches that worked when everything had a dedicated machine no longer work out both technically and in terms of human operational effort. This includes the dynamic nature of cloud-based environments where there can be high churn due to releases potentially several times an hour, heterogeneous performance of replicas and complex micro-service dependency graphs. There will be a quick overview of the history of the TSDB of Prometheus, giving context to the new storage engine in the 2.0 release and how it helps with the above technical problems. Following that will be a look at service discovery to find what to monitor in dynamic environments, the aggregation abilities of PromQL and how these can be combined to perform symptom rather than cause based alerting using techniques such as the RED method.

Monitoring can mean very different things to different people, and this often leads to confusion and misunderstandings. There are many offerings both free software and commercials, and it's not always clear where each fits in the bigger picture. This talk will look a bit at the history of monitoring, and then into the general categories of Metrics, Logs, Profiling and Distributed tracing and how each of these is important in Cloud-based environment.

Prometheus combines graphing and alerting in one package, with a powerful query language that lets you slice, dice, aggregate and predict what your system is going to do. Multi-dimensional labels are another core feature, allowing for a single alert definition to apply to thousands of dynamic hosts. Join us to understand the principles behind Prometheus and how you can use it to make monitoring your open-source project easier.