OpenTelemetry distributed tracing instead of logging everything

How many services are developed in your organization? And how many different routes can messages take between them? If the answer to the first question is "more than ten", then the answer to the second question is "more than a hundred", or even "more than a thousand".

Tracing processing errors or finding the source of distributed system slowdowns takes a lot of time. And the experience in using custom dashboards and logging indexes to solve these tasks is difficult to generalize to application outside a specific project. But all is not so hopeless. 

OpenTelemetry sets common standards for application metrics and trace export, allowing to aggregate them in a distributed environment on different platforms. And also visualize them using popular backends and dashboards - Datadog, Jaeger, Elastic/Kibana and Application Insights. 

The speaker will cover the basic principles of tracing in OpenTelemetry, discuss examples of use in web services, databases, queues and show how to export all this to different backends.