Code Monkey home page Code Monkey logo

Comments (2)

dominikh avatar dominikh commented on May 28, 2024

For now I'm going with the "recreating timelines" approach. Plenty of other tracers do this, too, so our users should be able to handle its inaccuracies.

We could also enrich our sample-based timelines with stack traces from runtime events. During a single 10ms span (i.e. one CPU sample), we may have access to many more stack traces of events that occurred during the sample, such as preemption, blocking/unblocking etc. This should probably be an opt-in feature, though. Turning a single CPU sample into a 10ms span is already making it tricky to interpret the data correctly. If we mix this with more accurate events then users are even more prone to assuming that the entire sampled timeline is accurate, or that a 9.8ms long span (the remainder of a sample) definitely was longer than a 0.2ms span (a span resulting from incorporating traced events).

Screenshot of the current WIP:
image

from gotraceui.

dominikh avatar dominikh commented on May 28, 2024

At the same time it looks like we might have to incorporate traced events. I'm still investigating, but at first glance it looks like we're not getting samples for goroutines that are blocked. This leads to the following screenshot:

image

I've disabled span merging so that each individual sample is visible. During the grey regions we're sleeping, and we don't seem to be receiving any samples, making it look like we're spending a lot of time in crypto/sha1.Sum, while in reality we've returned from it and are sleeping a lot.

from gotraceui.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.