Pixie
eBPF-based observability platform that gives instant visibility into Kubernetes applications without any code changes or manual instrumentation.
- Actively maintained
- Foundation-backed
- Well documented
- Permissive license
Overview
Pixie uses eBPF to automatically capture telemetry - HTTP/gRPC requests, database queries, network connections, resource usage - straight from the Linux kernel, so a team gets deep application observability without adding a single tracing library or agent to their code. Data stays in-cluster by default rather than being shipped externally.
Pixie is a fit for teams that want to debug a Kubernetes application right now without having pre-instrumented it, a fundamentally different approach from OpenTelemetry Collector's instrumentation-based model (also in this catalog).
Why it's in this catalog
Catalog rating: ★★★☆☆ (3/5)
eBPF-based auto-instrumenting observability for Kubernetes, a genuinely distinct no-code-change approach.
- Mature
- Foundation-governed
- Categories
- Monitoring
- Languages
- C++, Go
- License
- Apache-2.0
Related Projects
SkyWalking
Apache-2.0Application performance monitoring (APM) platform providing distributed tracing, service topology, and metrics in one system.
Actively maintained
- Monitoring
Kepler
Apache-2.0Kubernetes Efficient Power Level Exporter that measures and exposes the energy consumption of workloads running on a cluster.
Actively maintained
- Monitoring
Odigos
Apache-2.0Automatic OpenTelemetry instrumentation platform that adds distributed tracing to applications without changing a single line of code.
Actively maintained
- Monitoring
Jaeger
Apache-2.0CNCF distributed tracing platform for visualizing request flows across microservices, pinpointing exactly which service or call is responsible for latency or errors.
Actively maintained
- Monitoring
Spotted an error? Suggest an edit on GitHub.