• Home


Pixie’s long-term vision is to be the data plane for Kubernetes clusters. Pixie should make machine data such as application traffic and cluster performance, accessible and useful to anyone using Kubernetes. As a result, we want to build up the following high-level capabilities:

Expand Data Collection

Pixie currently provides rich, instant out-of-the-box visibility with basic metrics, infrastructure, and network traffic. We plan to expand our data collection to other forms of data, such as logs, full traces, and custom metrics. While we currently provide advanced code profiling for Go, C/C++, and Rust, we also plan to provide deeper support for other languages, such as Java, Ruby, and Python.

Improve Kubernetes Support

Pixie currently supports integration with basic Kubernetes resources, such as namespaces, services, and pods. We aim to expand this to other Kubernetes resources, including events.

Expand Edge ML/AI

Pixie’s edge compute engine allows us to apply ML/AI on unsampled data. We will expand on our applications of edge ML/AI, including detection of anomalous/interesting data, data compression, and more.

Support the ecosystem

Pixie has a versatile execution engine which can ingest and export data in a variety of formats. Pixie currently supports exporting data in the OpenTelemetry format which enables developers to consume Pixie data along with data produced by other tools (Jaeger, Prometheus). We plan to add ingest in the future. We also plan to leverage our client APIs in order to support tighter integrations with other open source projects.


We are building a data plane for Kubernetes rather than a full-fledged observability solution. Here are some of the items that we do not plan to tackle. Note that our API makes it possible to enable these use cases with downstream applications.

This site uses cookies to provide you with a better user experience. By using Pixie, you consent to our use of cookies.