Quickwit is a cloud-native search engine for observability built in Rust that handles logs, traces, and metrics workloads.
The tool addresses the challenge of searching and analyzing large volumes of observability data stored on cloud infrastructure. It decouples compute from storage, allowing stateless indexers and searchers to scale independently while querying data directly from cloud storage like Amazon S3, Azure Blob Storage, and Google Cloud Storage. This architecture enables sub-second search performance on cloud storage without requiring data to be loaded into memory-resident systems. The engine supports full-text search and aggregation queries through a Schemaless or strict schema indexing approach, and offers both schemaless analytics capabilities.
Quickwit suits teams building observability platforms who want to avoid the operational overhead of managing large in-memory search clusters. It works well for log management and distributed tracing use cases, with native support for OpenTelemetry, Jaeger, Kafka, Kinesis, and Pulsar as data sources. The tool provides an Elasticsearch-compatible API, allowing it to work with existing Elasticsearch and OpenSearch clients, and includes a Grafana data source plugin. Multi-tenancy, retention policies, and GDPR-compliant delete tasks make it suitable for enterprise deployments. Kubernetes integration through a Helm chart supports cloud-native deployment patterns.
The project maintains active development with regular feature releases and documentation updates. The codebase shows consistent refinement of core search and indexing capabilities. Community engagement is supported through Discord channels and comprehensive tutorials covering practical scenarios like trace analytics with Grafana and distributed tracing with Jaeger. The project roadmap explicitly includes metrics support as an upcoming feature, indicating planned expansion beyond current log and trace functionality.
