DevOps Articles

Why We Built Cloud Native Object Storage for PostgreSQL with Amazon S3

It’s no secret that at https://www.timescale.com/?utm_source=new-stack&utm_medium=web&utm_campaign=data-tiering-private&utm_content=timescale-homepage, we’re fans of PostgreSQL — we offer high-performance PostgreSQL in the cloud. And current cloud database solutions don’t give time series developers the tools they need to address this problem.

So to solve this problem of cost-efficient storage of time-series data, we decided to build what our users would want to build themselves: we added a bottomless, consumption-based object store based on Amazon S3 to our cloud native PostgreSQL databases.

This allows developers to scale their time series data infinitely and cost-effectively — keeping some data in the database itself and tiering data to object storage to save on costs.

SELECT add_tiering_policy ('metrics', INTERVAL '2 weeks'); After tiering the data to the object store, developers can continue to query it from within the database via standard SQL.