Quickwit

Quickwit is a cloud-native, open-source search and analytics engine built in Rust, designed as a cost-effective alternative to Elasticsearch, Splunk, Datadog, and similar observability platforms.

Reviewed by 7wData

On this page

Publisher review

Quickwit is a cloud-native, open-source search and analytics engine built in Rust, designed as a cost-effective alternative to Elasticsearch, Splunk, Datadog, and similar observability platforms. It executes complex search and aggregation queries directly on cloud object storage—S3, Azure Blob Storage, MinIO—with sub-second latency, powered by the Tantivy search library. The platform uses a decoupled compute-storage architecture that allows independent scaling of indexing and search workloads, enabling teams to spin up clusters in seconds without data migration.

Quickwit is optimized for log management, distributed tracing, and time-series analytics, with native support for OpenTelemetry, Jaeger, and Prometheus labeling patterns. It supports both schemaless JSON indexing and strict schema enforcement, making it adaptable to diverse observability use cases. Real-world deployments report 5–20x reductions in compute requirements and 2x storage savings compared to traditional search engines.

Datadog acquired Quickwit in January 2025, transitioning its license from AGPL to Apache 2.0. The platform remains actively maintained with 11,319 GitHub stars and is particularly suited for organizations needing cost-efficient log retention at scale, edge-case observability workloads, and deployments requiring data residency or regulatory compliance constraints.

Get the AI & data signal, daily.

335k+ subscribers read this every morning. One email, both newsletters. Unsubscribe anytime.

How it works

  1. Sub-second search on object storage

    Executes full-text and aggregation queries directly on S3, Azure Blob, or compatible storage with minimal latency via optimized indexing and vectorized processing.

  2. Decoupled compute and storage

    Stateless indexers and searchers scale independently; data stays in cheap object storage while compute spins up or down on demand, eliminating data reshuffling.

  3. OpenTelemetry and Jaeger native support

    Schemaless ingestion of OTEL traces and Jaeger spans; compatible with Prometheus-style label indexing for observability workflows.

  4. Elasticsearch query language compatibility

    Accepts Elasticsearch Query DSL and REST API patterns, lowering migration friction for teams moving off Elasticsearch or Splunk.

  5. Kubernetes and serverless deployment

    Supports single-node, multi-node on-prem, and Kubernetes with Helm; also runs on AWS Lambda for fully serverless log ingestion.

  6. Multi-tenancy and retention policies

    Indexes multiple partitions and datasets in one cluster; GDPR-compliant deletion, tiered retention, and isolation per logical tenant.

  7. Cost benchmarking (10x cheaper claims)

    Published and field-tested claims of 10x cost efficiency over Elasticsearch, Splunk, and Datadog for comparable log volume and retention.

Strengths and trade-offs

Strengths

  • Exceptional cost-per-GB and CPU efficiency; real deployments report 5–20x compute reductions compared to Elasticsearch or Splunk.
  • Decoupled architecture means no need to resize clusters or move data; scales compute and storage independently, ideal for variable workloads.
  • Low barrier to entry: fully open-source (Apache 2.0), self-hosted without licensing, and Elasticsearch DSL compatibility cuts migration effort.

Trade-offs

  • Indexing speed and CPU consumption lag Elasticsearch; benchmarks show ~80% higher CPU burn during ingest, making it suboptimal for extremely high-volume real-time pipelines.
  • Smaller ecosystem and community (11k GitHub stars vs. Elasticsearch's 76k+); fewer integrations, plugins, and reference architectures for edge cases.
  • Acquisition by Datadog in January 2025 creates vendor lock-in risk; unclear whether OSS version receives feature parity or is deprioritized relative to Datadog's commercial offering.

Pricing context

Quickwit is fully open-source and free under Apache 2.0 for self-hosted deployment, with no licensing costs. Managed hosting via third-party platforms (e.g., Elestio) starts at $14/month for minimal resources, billed hourly based on compute and storage usage. Users pay only for cloud infrastructure (S3, MinIO, or Azure storage) when running their own instances. No official Quickwit-hosted SaaS pricing; Datadog acquisition may lead to commercial offerings, but as of mid-2026 none are publicly available.

Alternatives

User reviews

No user reviews yet. Be the first to write one.

Sources

Reporting on this tool draws on these publicly available sources.

  1. quickwit.io — Core features, deployment options, cloud-native architecture, observability use cases
  2. quickwit.io — Architecture, design principles, schemaless indexing, OpenTelemetry/Jaeger support, decoupled compute and storage
  3. github.com — Project status, active development, GitHub stars (11,319), open-source license, recent updates (June 2026)
  4. www.computerweekly.com — Cost efficiency claims (10x cheaper), target market, competitive advantages, real-world customer results
  5. www.crunchbase.com — Datadog acquisition date (January 9, 2025), acquisition context and strategic rationale
  6. quickwit.io — License change from AGPL to Apache 2.0, acquisition announcement and implications
  7. openalternative.co — Competitive comparison with Elasticsearch, architectural differences, performance trade-offs
  8. elest.io — Managed hosting pricing ($14/month), open-source licensing, infrastructure cost model