NEW SnappyData takes on Aerospike: a Performance Benchmark

SnappyData, the Apache Spark Database. Stream - Transact - Analyze - Predict all in one cluster

SnappyData is a high performance in-memory data platform for mixed workload applications. Built on Apache Spark, SnappyData provides a unified programming model for streaming, transactions, machine learning and SQL Analytics in a single cluster. SnappyData unifies real time data sources with external data sources that have a Spark connector. SnappyData is 100% compatible with Apache Spark and is Open Sourced with an Apache V2 license.


SnappyData delivers a 20X speed-up to Spark and a 100X speed-up using synopses data

Open Source

100% compatible with Apache Spark. 100% committed to Open source.




Easily work with JSON/SQL/Objects. Work with all data sources supported by Spark.


One cluster to support transactions, streaming and analytics. Simpler to deploy, manage and own

How industries use it

IoT, Health, Ad Tech & More

  • Industrial Internet

    Continuous ingestion of sensor data analyzed to predict failures before they happen.

  • Finance

    Analytics Platform as a service in Finance: post-trade and tick analytics.

  • Ad Tech

    Exploratory analytics for data scientists in the Ad Tech world

Explore SnappyData's Vision

SnappyData's CIDR 2017 paper is available