# Databricks makes multi-table transactions generally available

_Published Friday, October 16, 2026 at 6:08 PM EDT · Infrastructure, Products · Latest · Tier 2 — Notable_

![Databricks makes multi-table transactions generally available — Primary](https://www.databricks.com/sites/default/files/2026-10/begin-transaction-commit.png)

Databricks announced general availability of transactions that let multiple database statements across multiple tables succeed or fail together. The capability supports Delta Lake and Apache Iceberg tables managed through Unity Catalog, allowing teams to run multi-step warehouse workflows directly on their lakehouse data platform.

Transactions can also write to streaming tables and apply row-level security and column masking to reads and writes. Fivetran says it now commits the two writes needed to replace a historical record as one operation, removing consistency-repair code. Concurrent transactions can write to the same table provided they do not update the same row.

## Sources

- [Databricks](https://www.databricks.com/blog/transactions-are-generally-available-advancing-warehousing-lakehouse)

---
Canonical: https://techandbusiness.org/newswire/uH5uwcBG8kgereffwtm3cd
Published: 2026-10-16T22:08:14.362Z
Story chronology: 2026-10-16T21:00:00.000Z
Retrieved: 2026-10-17T00:53:24.889Z
Publisher: Tech & Business (techandbusiness.org)
