Open table format that adds ACID transactions to lakehouse tables across Spark, Trino, and more
Apache-2.0
- Scala
- Java
- Python

About Delta Lake
Delta Lake is an open table format for building lakehouse architectures. It layers ACID transactions and a versioned transaction log over object storage, so data teams get reliable inserts, updates, and concurrent access on tables that compute engines like Spark, PrestoDB, Flink, Trino, and Hive can read and write.
The format ships with APIs for Scala, Java, Rust, Ruby, and Python. Spark applications work through standard read, write, readStream, and writeStream calls, while Delta Standalone gives single-node JVM apps direct access to table metadata and the transaction log without Spark.
Delta Lake guarantees serializability for concurrent reads and writes and keeps backward compatibility, so newer releases can always read tables written by older ones.
Key features
- ACID transactions over object storage
- Read and write from Spark, PrestoDB, Flink, Trino, and Hive
- APIs for Scala, Java, Rust, Ruby, and Python
- Delta Standalone for single-node JVM access without Spark
- Serializable concurrent reads and writes
Details
- First released
- 2019
- Type
- Open table format
- Language
- Scala · Java · Python
- Compatibility
- Spark · Trino · Flink · PrestoDB · Hive
- Storage
- Delta tables on object storage
- Governance
- Linux Foundation
