YDB
activeby YDB (Yandex) · since 2022 · also known as Yandex Database, YDB Platform
Open-source distributed SQL from Yandex combining row and column tables, topics and vector search.
Overview
YDB is a distributed SQL database that Yandex built for its own services and open-sourced under Apache 2.0 in 2022. It provides serializable transactions over automatically split, synchronously replicated row tables, plus column-oriented tables for analytics, persistent topics with a Kafka-compatible API, and vector indexes. Clusters can span three availability zones and survive the loss of a zone, and the same binary scales from a single node to thousands of servers. Queries use YQL, a SQL dialect, with a PostgreSQL-compatibility layer in development. It is a strong but under-marketed alternative to CockroachDB or TiDB, best known in Russia and among Yandex Cloud users.
Strengths & trade-offs
Strengths
- +Serializable distributed transactions with automatic sharding
- +Row tables, column tables and persistent topics in one system
- +Kafka-compatible topics with transactional writes alongside tables
- +Survives full availability-zone loss in three-AZ mode
- +Permissive Apache 2.0 license
Trade-offs
- −YQL is not standard SQL; PostgreSQL compatibility is still experimental
- −Smaller international community; much content is Russian-first
- −No triggers or stored procedures
- −Operational tooling less mature outside Yandex Cloud
Great for
- +High-load OLTP services needing automatic sharding
- +Combined transactional and analytical workloads on one platform
- +Event streaming with transactional consistency to tables
- +Serverless backends on Yandex Cloud
Think twice for
- −Teams needing a drop-in PostgreSQL or MySQL replacement
- −Small single-node applications
- −Stored-procedure-heavy legacy apps
Capabilities
Native or official first-party support · 15 of 25 tracked
Scale & consistency
- Comfortable data size
- Massive · Multi-PB
- Horizontal scaling
- Native
- Multi-region
- Active active
- Transactions
- Full ACID
- Consistency
- Strong, Tunable
- Schema
- Strict
- Architecture
- Distributed, Client server, Serverless
- Query languages
- YQL (SQL dialect)
Automatic table splitting by load and size; three-AZ (mirror-3-dc) clusters tolerate zone failure; runs thousands of nodes inside Yandex.
Deployment
Ecosystem
Alternatives to YDB
Postgres-wire distributed SQL built to survive node, zone and region failures with serializable transactions.
ydb vs cockroachdb →MySQL-compatible distributed SQL with a row store for OLTP and a columnar replica for real-time analytics.
ydb vs tidb →Apache-licensed distributed SQL that reuses the PostgreSQL query layer on a Raft-replicated storage engine.
ydb vs yugabytedb →Google's globally consistent, horizontally scaling SQL service with graph, vector and full-text built in.
ydb vs google-cloud-spanner →Apple's ordered, strictly serializable distributed key-value store, built to have layers on top.
ydb vs foundationdb →Open-source columnar OLAP database known for very fast aggregations over billions of rows.
ydb vs clickhouse →YDB FAQ
What is YDB?
YDB is a relational database by YDB (Yandex), first released in 2022. YDB is a distributed SQL database that Yandex built for its own services and open-sourced under Apache 2.0 in 2022. It provides serializable transactions over automatically split, synchronously replicated row tables, plus column-oriented tables for analytics, persistent topics with a Kafka-compatible API, and vector indexes. Clusters can span three availability zones and survive the loss of a zone, and the same binary scales from a single node to thousands of servers. Queries use YQL, a SQL dialect, with a PostgreSQL-compatibility layer in development. It is a strong but under-marketed alternative to CockroachDB or TiDB, best known in Russia and among Yandex Cloud users.
Is YDB open source?
Yes. YDB is open source. License: Apache License 2.0.
Does YDB support ACID transactions?
Yes. YDB supports full ACID transactions, with strong, tunable consistency.
Does YDB scale horizontally?
YDB scales horizontally out of the box, with built-in sharding or partitioning. It is comfortable at massive scale (Multi-PB). It supports active-active multi-region writes.
How can I run YDB?
You can self-host it, use the vendor's managed cloud and run it serverless. Managed offerings include Managed Service for YDB.
What is YDB best for?
YDB is a strong fit for high-load OLTP services needing automatic sharding; combined transactional and analytical workloads on one platform; and event streaming with transactional consistency to tables. Think twice for teams needing a drop-in PostgreSQL or MySQL replacement.
What are the best alternatives to YDB?
Common alternatives to YDB include CockroachDB, TiDB, YugabyteDB, Google Cloud Spanner and FoundationDB. The right choice depends on your workload, scale and hosting; WhichDB's smart match ranks them against your requirements.
How popular is YDB?
YDB ranks #103 of 844 databases on the WhichDB Score (27.4/100), and #32 among relational databases. The score combines public adoption signals such as package downloads, GitHub activity, job posts and developer discussion.