Time Series Database Design Techniques

Distributed SQL best practices for fast, scalable, always-on time series

Webinar Description

If you are monitoring or analyzing time-stamped data like sensor readings, IT system metrics, GPS locations, logs, this webinar will explain how to build a scalable, flexible database to manage them.

Using Postgres-compatible CrateDB, we will cover the following time series database design best-practices:

  • Data partitioning
  • Data sharding
  • Data replication (for HA)
  • Data aging/archiving
  • Handling dynamic structures (JSON)
  • Extending/changing schemas


45 minutes

What you will learn

  • Distributed SQL database design basics - partitions, sharding, replicas

  • Scaling time series to handle more data, more users, more queries

  • SQL vs. NoSQL DB options for time series

Watch On-Demand

About the Speakers

Andy Ellicott

Andy Ellicott, Product & Developer Relations


A former software engineer, Andy has spent the last 20 years launching and evangelizing new databases. Before joining Crate.io in 2016, Andy helped launch the Cloudant DBaaS (acquired by IBM), and before that, launched VoltDB, Vertica, and ObjectStore (last millennium!).


Yash, Field Engineer


Yash Lanka is based in New York City and heads up Crate.io field engineering activity in the US. Prior to joining Crate in 2017, Yash has years of experience designing and developing distributed systems in IoT and other fields, with Cap Gemini, GE, and IGATE.