# lakehq/sail

Drop-in Apache Spark replacement written in Rust, unifying batch processing, stream processing, and compute-intensive AI workloads.

Repository: https://github.com/lakehq/sail
Canonical: https://ross.abutalabs.com/products/sail
Homepage: https://lakesail.com
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: arrow, big-data, pyspark, rust, spark, sql, datafusion, python, artificial-intelligence, data-engineering, distributed-computing, machine-learning, lakehouse, delta-lake, iceberg, spark-connect, spark-sql, apache-iceberg, apache-spark, parquet
Last push: 2026-08-26T16:21:26+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 70
- inputs: {"age_days": 986, "days_push": 7, "days_rel": 9, "gap_med": 14, "n_releases_24m": 28}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3333, forks 210 (observed 2026-08-28T04:07:56.676151+00:00)

## What it is
Sail is an open-source, Rust-native multimodal compute engine that serves as a drop-in replacement for Apache Spark, unifying batch processing, stream processing, and compute-intensive AI workloads. It is compatible with the Spark Connect protocol, so existing PySpark and Spark SQL code runs unchanged, with no JVM overhead, and supports Delta Lake and Apache Iceberg lakehouse formats.

## Use cases
- replace Apache Spark with a faster Rust engine without rewriting PySpark code
- run Spark SQL and DataFrame workloads with no JVM overhead
- unify batch and stream processing on a single compute engine
- run compute-intensive AI and machine learning workloads on a lakehouse
- query Delta Lake and Apache Iceberg tables natively
- deploy distributed data processing on Kubernetes
- cut infrastructure costs for large-scale Spark deployments

## When to choose
- you have existing PySpark or Spark SQL code and want better performance without rewrites
- you want to avoid JVM tuning, slow startup, and memory overhead
- you need a lakehouse engine supporting Delta Lake and Iceberg
- you want to run batch, streaming, and AI workloads on one engine
- you want to self-host a Spark-compatible engine or deploy on Kubernetes

## When to avoid
- you rely on Spark features or third-party libraries not yet supported by Sail's compatibility layer
- you need the full mature ecosystem and long-term battle-tested stability of Apache Spark
- your workloads depend on JVM-specific integrations or custom Scala/Java UDFs
- you need guaranteed enterprise support without a commercial plan

## Facets
- artifact type: framework
- maturity: active
- function: etl, streaming, machine-learning, data-science
- domain: big-data, microservices, machine-learning, artificial-intelligence
- platform: rust, python, cloud, self-hosted, cli
- tags: apache-spark, spark-connect, pyspark, datafusion, arrow, delta-lake, apache-iceberg, lakehouse, parquet, distributed-computing, batch-processing, stream-processing, no-jvm, tpc-h, clickbench, sql, data-engineering, kubernetes, docker

## Member repositories
- lakehq/sail (main) score 93

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:56.676151+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:41:41.552174+00:00, confidence not recorded.
  - readme: https://github.com/lakehq/sail (fetched 2026-08-28T04:07:56.676151+00:00, sha a3d5abef5703)
  - homepage: https://lakesail.com (fetched 2026-08-29T09:34:38.081833+00:00, sha 8b933ba724c3)
  - site_page: https://docs.lakesail.com (fetched 2026-08-29T09:34:38.093090+00:00, sha 8bafaca71566)
  - site_page: https://docs.lakesail.com/platform (fetched 2026-08-29T09:34:38.094878+00:00, sha d9fc6bf28231)
  - site_page: https://lakesail.com/about (fetched 2026-08-29T09:34:38.096465+00:00, sha 0c9e6e3d26a1)
  - site_page: https://lakesail.com/pricing (fetched 2026-08-29T09:34:38.091060+00:00, sha 320d198ed7aa)
- Data as of 2026-08-30T08:39:29.467469+00:00.