Cardinal Lakerunner
Turn your S3-compatible object store into a blazing-fast observability backend
What is Lakerunner?
Lakerunner is an event-driven ingestion engine that transforms S3-compatible object storage into a real-time observability backend. It automatically monitors for structured telemetry data, converts it into optimized Apache Parquet format, and handles indexing, aggregation, and compaction operations in real time—all without vendor lock-in.
⚡Event-Driven Architecture
Triggers ingestion automatically when new objects are created in S3 buckets via S3 Object Notifications.
🔓No Vendor Lock-In
Works with any S3-compatible object storage. Your data stays in your infrastructure.
Key Features
Automatic S3 Ingestion
Triggered by object creation events for real-time data processing.
OpenTelemetry Native
Native support for OpenTelemetry log and metric proto files.
Multiple Data Sources
CSV, JSON.gz, DataDog, FluentBit, and more.
Quick Setup
Deploy locally in under 5 minutes with automated installation.
Grafana Integration
Built-in Grafana plugin for immediate data visualization and querying.
Cloud Native
Docker and Kubernetes support with Helm charts.
Supported Input Formats
OpenTelemetry Protocol Files
Native support for OTLP logs and metrics
CSV Data
Structured data from CSV files
JSON.gz Compressed Files
Compressed JSON data ingestion
DataDog & FluentBit
Integration with popular observability tools
Getting Started
Deploy Lakerunner locally in under 5 minutes with our automated installation script
Requirements
- ✓Docker
- ✓Kubernetes (via kind or minikube)
- ✓kubectl
- ✓Helm 3.14+
Once running: Access Grafana at localhost:3000 to explore your ingested telemetry through the bundled Cardinal datasource plugin.