Time Series Service Overview

About the Time Series Service

Time Series data is a sequence of data points collected at set time intervals over a continuous period of time. Sensor data is an example of a common way to generate time series data. A Time Series data store requires a measurement with a corresponding timestamp. The Time Series service provides an attributes field to include additional relevant details about that specific data point, such as units or site, for example, "Site":"SanFrancisco".

Time Series data can consist of regular data (data sampled at regular time intervals), or irregular data, for example, data that is recorded only when a certain event occurs (so always at random times).

Note: Time Series is not an archival service. The time to live (TTL) period for data stored in the Time Series service is two years from the date of ingestion, after which the data will be deleted. You should back up your data if you need to keep it longer.

The Time Series service provides the following benefits:

  • Efficient storage of time series data.
  • Indexing the data for quick retrieval.
  • High availability so you can access and query your data from anywhere via HTTP.
  • Horizontal scalability.
  • Millisecond data point precision.
Component Description
Data IngestionThe data-ingestion layer provides the ability to ingest real-time streaming data.
Data QueryThe query API allows you to query data, with support for grouping of data points by tags, time ranges, or values, as well as aggregations. You can also filter by attributes to narrow your results.

The below high-level architecture diagram shows different ways you can send data to the Time Series service for ingestion, as well as how users can interact with the query service.

Figure: Time Series Service Architecture
The service ingests data through a WebSocket connection, which can then be queried using a REST API with HTTP requests. The below are some example ingestion configurations:
  • Site A – Historian

    Sensors produce time series data associated with a tag name and send it to Historian for storage and management. S2C subscribes to tags and collects generated data from those tags only. S2C requests a web socket connection from a gateway application that is used for data ingestion.

  • Site B – WebSocket River utility

    The WebSocket River establishes a connection when you first attempt a data transfer and keeps that channel open for as long as possible. Each data transfer verifies that the websocket connection is open. If the connection has been closed, the service opens a new connection. For more information about WebSocket River, see WebSocket River.

    To communicate with the Time Series gateway, the data must be structured as shown in tss-using-service.html#concept_0449dd2d-724f-4fd1-b49c-038886b155a5.

  • Site C – Direct WebSocket connection

    Devices use an application to communicate directly with the websocket.

Additional Information

Exploring Time Series Tutorials