Amazon Timestream is a fully managed database service designed for IoT, industrial telemetry, and similar operational applications that require storing trillions of events per day.

 

What is Time-series data?

As the name implies Time-series data has Time and Series (sequence) as of key significance elements of the overall data generated. Following are some key characteristics of this Time-series data:

  • Order (with respect to Time)
  • Typically is append-only data (variation from recently received data)
  • Data over certain time-intervals have more relevance – compared to data spread far apart

 

Key Points on Amazon Timestream

  • Amazon Timestream is a fully managed service.
  • Amazon Timestream is purpose-built to efficiently store and process time-series data.
  • Amazon Timestream automates rollups, retention, tiering, and compression of data.
  • Amazon Timestream handles time-series data 1,000 times faster than a typical relational database.
  • Amazon Timestream is much economical for handling times-series data.
  • Has built-in functions – such as smoothing, approximation, and interpolation – to facilitate efficient analytics for time-series data.
  • Uses Cases – IoT applications, DevOps, Application Monitoring, Industrial Telemetry

 

Following diagram shows how Amazon Timestream works:

How Amazon Timestream works

Image courtesy of AWS

 


Pricing

Amazon Timestream is charged based on following components:

  • Writes – per 1 million writes of 1 KB size
  • Query – per TB of data scanned by each query
  • Storage – per GB
    • Different rates depending on storage type – Memory Store, SSD Store, Magnetic Store
  • Data Transfer – standard data transfer rate applies when data is transferred out of AWS