Origins & Support

KalDB began with production search work at Slack and continued through related development at Airbnb. KalDB is a fork that brings these efforts together into an open source log search and analytics system for high-volume observability workloads.

The project keeps the core architectural ideas that made Astra practical at scale: durable ingest through Kafka, Lucene-based indexing, object storage for long-term indexed data, and independently scalable compute tiers for ingest and query serving.

Additional KalDB capabilities include:

• Admin UI for cluster inspection and management.
• OpenTelemetry support for modern observability pipelines.
• Enterprise features and company-backed support for production deployments.
• 100% support for OpenSearch analytics queries, including ClickBench coverage.
• Broader aggregation support, including multi_terms, sibling aggregations, bucket sorting, and bucket filtering.
• Improved Grafana and Zipkin-compatible trace workflows.
• Expanded operational documentation for configuration, monitoring, metrics, troubleshooting, and recovery.