To get insights into the performance of your application and what specifically it’s spending time on, you need good tracing tools.
Historically however, it’s been very difficult to implement tracers in Node.js due to its async nature and the lack of an official context propagation API in core.
Throw modern deployment architectures into the mix, and you also need to deal with distributed tracing across multiple microservices.
This talk is about how tracers work behind the scenes, which pitfalls exist and the work going on in the Node.js Diagnostics Working Group, in W3C, and within other standard-organizations to improve the situation.