# ytsaurus/ytsaurus

YTsaurus is a scalable and fault-tolerant open-source big data platform.

Repository: https://github.com/ytsaurus/ytsaurus
Canonical: https://ross.abutalabs.com/products/ytsaurus
Homepage: https://ytsaurus.tech
Language: C++
License: Apache-2.0
License Family: permissive
Topics: big-data, clickhouse, distributed-database, lakehouse, olap-database, spark, sql, ytsaurus
Last push: 2026-08-26T21:23:08+00:00

## Health v2 (maintenance only)
Score: 94/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 86, longevity 97
- inputs: {"age_days": 1367, "days_push": 7, "days_rel": 19, "gap_med": 4, "n_releases_24m": 102}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2200, forks 218 (observed 2026-08-28T04:06:25.597013+00:00)

## What it is
YTsaurus is an open-source, fault-tolerant big data platform combining distributed storage, MapReduce processing, a SQL query engine, and a NoSQL key-value store. It integrates ClickHouse (CHYT) for fast analytics and Apache Spark (SPYT) for ETL, and scales to exabytes of data and millions of CPU cores.

## Use cases
- run MapReduce batch jobs over large datasets
- run ad hoc SQL analytics without exporting data to an external OLAP system
- store and query data with a low-latency transactional key-value store
- build ETL pipelines with Apache Spark or SQL
- manage GPU clusters for training large machine learning models
- store transactional metadata for distributed systems
- deploy a multi-tenant big data platform on Kubernetes

## When to choose
- you need a scalable, fault-tolerant platform combining storage, batch processing, and analytics in one system
- you want multi-tenant big data infrastructure serving many users on shared hardware
- you need exabyte-scale storage with no single point of failure
- you want ClickHouse-style SQL analytics directly on your data lake
- you need a transactional key-value store for OLTP workloads alongside batch processing

## When to avoid
- you only need a simple single-node database or analytics engine
- you want a fully managed cloud service without operating your own cluster
- your workloads are small enough for Postgres, ClickHouse, or Spark alone
- you need a lightweight embedded or edge deployment

## Facets
- artifact type: service
- maturity: active
- function: database, search-engine, etl, streaming, scheduling, caching
- domain: big-data, databases, microservices, analytics
- platform: self-hosted, cloud
- tags: big-data, mapreduce, distributed-file-system, nosql, olap, sql-query-engine, clickhouse, apache-spark, key-value-store, data-lakehouse, data-engineering, linux, docker, kubernetes

## Member repositories
- ytsaurus/ytsaurus (main) score 94

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:25.597013+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 2026-08-30T02:47:06.362719+00:00, confidence not recorded.
  - readme: https://github.com/ytsaurus/ytsaurus (fetched 2026-08-28T04:06:25.597013+00:00, sha 8bcfe4f072bf)
  - homepage: https://ytsaurus.tech (fetched 2026-08-29T10:27:19.704247+00:00, sha 7e1c3cf39411)
  - site_page: https://ytsaurus.tech/docs/en (fetched 2026-08-29T10:27:19.713275+00:00, sha e9a813b24cd6)
  - site_page: https://ytsaurus.tech/docs/en/overview/try-yt (fetched 2026-08-29T10:27:19.715370+00:00, sha 8c58b14cdfe4)
- Data as of 2026-08-30T08:39:29.467469+00:00.