Your business generates a lot of data every second. Whether it's in-app, in-store or on-line transactions, stock trades, sensor readings, or social networking interactions, you want to analyze that data so you can understand what's happening right now.
This creates challenges for your database management system:
These challenges are hard for databases. Traditionally, it's difficult to get good data insert and fast query performance at the same time from the same database, especially if there are lots of people running queries or performing ad-hoc analysis. It's harder to scale the database over time to handle more data and more queries. Integrating non-relational data into analytics is also challenging, requiring the use of multiple databases and ETL code to integrate them.
Crate is an open source SQL database designed to meet these challenges.
Fast INSERT performance - Crate can concurrently ingest hundreds of thousands of structured or unstructured records per second on multiple tables. This makes it ideal for time-series data, sensor data, or log data.
Fast, ad-hoc querying - Analytic dashboards often aggregate data and enable users to explore results by drilling down into reports and visualizations. Crate's distributed, columnar architecture enables many concurrent users to explore data like this with fast response times, even when querying billions of rows of data.
Advanced analytics - Crate is a multi-model database that can query structured and unstructured data together. This means you can create geospatial, full-text, time-series, and other interesting analytical results.
Lower-costs - Crate is open-source and free for you to use. It's designed to run fast and scale on inexpensive commodity hardware. Customers have reported that switching to Crate has reduced the size of their database platforms by over a half.
Easy to change - Crate is designed for change. You can add columns to support new data items and run arbitrary queries to create new insights with no performance penalty.
Works with any SQL tools - Crate is accessed via SQL through JDBC. This enables it to be accessed by business intelligence reporting, data visualization, data integration and pipelining tools your team enjoys using today.
Simple scalability - Crate is masterless and fully distributed so can run on an elastic compute cluster. Through the use of container and automatic data redistribution technology, Crate makes it easy to scale out to increase performance or storage by adding new container nodes to your cluster.