Skip to content
master
Switch branches/tags
Code

Talk: Observing Distributed Systems

Presented at:

Observability is the ability to understand what is going on with your systems, not only from the point of view of how the system looks from outside, but been able to answer more granular questions, like where did this message goes?

Metrics, Logging and Tracing are called the three pillars of Observability.

In this presentation we will see how we can use these tools and how they are related to be able to observe our systems.

  • Logging and Metrics

  • OpenTracing API

  • Demo: Tweets App

Tools

  • JDK 8

  • Docker (Docker-Machine, host: docker-vm)

  • Logging: Fluentd, Elasticsearch Kibana

  • Metrics: Prometheus

  • Tracing: OpenTracing, Jaeger, Zipkin

  • Frameworks/Libraries: Dropwizard, JOOQ, Kafka Clients, HTTP Client, Elasticsearch, Postgresql.

Key takeaways

  • Distributed Tracing is just one more tool for your toolkit, and is not mean to replace metrics and logging, and it could be seen as an abstraction of them.

  • OpenTracing is an effort to standarize how to instrument your applications, so you can upgrade/migrate your infrastructure without changing your implementations.

  • OpenTracing is a young project, go ahead, try out and give feedback to the community, or contribute to make it better.

Resources

Papers:

Blog posts:

talks

Books:

About

No description or website provided.

Topics

Resources

License

Releases

No releases published

Packages

No packages published