ytsaurus
YTsaurus is an open-source distributed big data platform that combines storage, processing, and analytics capabilities. It supports MapReduce, SQL queries (via ClickHouse integration), Spark jobs, and key-value store operations across massive deployments.
Key facts
Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.
| Field | Value |
|---|---|
| Repository | ytsaurus/ytsaurus |
| Owner | ytsaurus |
| Primary language | C++ |
| License | Apache-2.0 — OSI-approved |
| Stars | 2.2k |
| Forks | 208 |
| Open issues | 487 |
| Latest release | docker/ytsaurus/25.3.1 (2026-04-02) |
| Last updated | 2026-07-08 |
| Source | https://github.com/ytsaurus/ytsaurus |
What ytsaurus is
C++ distributed system providing multi-tenant data processing via MapReduce, CHYT (ClickHouse SQL engine), SPYT (Spark integration), ACID transactions, and a NoSQL key-value store. Designed for fault tolerance, horizontal scaling to exabyte-scale, and resource isolation across thousands of nodes.
Get the ytsaurus source
Clone the repository and explore it locally.
git clone https://github.com/ytsaurus/ytsaurus.gitcd ytsaurus# follow the project's README for install & configurationNeed it deployed, integrated, or customized instead? DEV.co ships production installs.
Best use cases
Implementation considerations
- Plan for persistent storage backend (HDD/SSD/NVME) and redundancy model early; replication overhead scales with cluster size and data volume.
- Cluster bootstrap requires careful capacity planning (CPU, memory, network) and monitoring infrastructure to manage thousands of potential nodes.
- MapReduce, CHYT, and SPYT subsystems have distinct tuning parameters and failure modes; resource isolation policies must be defined before production traffic.
- Schema design and partitioning strategy significantly impact query performance; leverage ClickHouse SQL documentation for CHYT specifics.
- Upgrade and maintenance windows involve coordinated state transitions; understand rolling update semantics to prevent data loss.
When to avoid it — and what to weigh
- Small or prototype deployments — Operational overhead of cluster management, replication, and multi-node coordination is justified only for substantial data volumes and distributed workloads.
- Simple OLTP applications — While YTsaurus includes a key-value store, traditional RDBMS products (PostgreSQL, MySQL) are simpler to operate for standard transactional workloads.
- Teams with no C++ or distributed systems expertise — Building, debugging, and troubleshooting the platform requires deep infrastructure knowledge; production deployments demand experienced DevOps and platform engineering staff.
- Rapid time-to-market requirements — Learning curve, cluster provisioning, and integration with existing tools (Spark, ClickHouse) extend implementation timelines compared to managed SaaS alternatives.
License & commercial use
Licensed under Apache License 2.0, a permissive OSI-approved license. Source code, modifications, and derivative distributions are permitted under standard Apache 2.0 terms.
Apache 2.0 permits commercial use, modification, and distribution. However, verify compliance with any third-party dependencies (ClickHouse, Spark, etc.) and review patent clauses in Apache 2.0 for your jurisdiction and use case. No warranty is provided; production deployments require your own SLA and support strategy.
DEV.co evaluation signals
Editorial assessment — not user reviews. Directional, with an explicit confidence level.
| Signal | Assessment |
|---|---|
| Maintenance | Active |
| Documentation | Adequate |
| License clarity | Clear |
| Deployment complexity | High |
| DEV.co fit | Good |
| Assessment confidence | High |
No specific security audit, vulnerability disclosure policy, or threat model data provided. Considerations: distributed system with multi-tenant workloads requires careful authentication (user isolation), authorization (resource quotas), encryption in transit/at-rest (verify implementation), and audit logging. Review security documentation and conduct threat modeling before exposing to untrusted tenants or sensitive data.
Alternatives to consider
Apache Hadoop / Apache Spark clusters
Mature, widely-deployed MapReduce and Spark ecosystems; simpler operational model for many organizations, but less integrated SQL/OLAP layer and no built-in key-value store.
Databricks (managed Spark)
Hosted, fully managed Spark alternative with integrated notebooks and Delta Lake; higher SaaS cost but eliminates cluster operational burden and provides vendor support.
Snowflake / BigQuery
Cloud-native analytics warehouses with SQL-first model, automatic scaling, and zero cluster management; suitable for organizations prioritizing simplicity over platform control.
Build on ytsaurus with DEV.co software developers
Contact our engineering team to assess architecture fit, deployment complexity, and integration strategy for your scale and team expertise.
Talk to DEV.coRelated on DEV.co
Explore the category and the services that help you build with it.
ytsaurus FAQ
Is YTsaurus production-ready?
What is the difference between CHYT and SPYT?
Can I run YTsaurus on managed Kubernetes?
What is the learning curve for teams familiar with Hadoop or Spark?
Software development & web development with DEV.co
Need help beyond evaluating ytsaurus? DEV.co is a software development agency offering software development services and web development for teams of every size. Our software developers and web developers build custom software, web applications, APIs, and open-source databases integrations — and maintain them long-term.
Ready to evaluate YTsaurus for your data infrastructure?
Contact our engineering team to assess architecture fit, deployment complexity, and integration strategy for your scale and team expertise.