Understand Links

Use links to read and write data from different data sources. Links associate an analytics object with a specific data source.

Each link is associated with a specific type of data element represented by an analytics object. Multiple links can share the same analytics object if the data they access has the same structure. For example, you can read an object from Oracle Database Cloud Service and then write it to .

In your analytics processors you can read and write data using links. Links aren’t specific to an analytics processor, you can use the same link from different analytics processors.

When you define an analytics processor you must define the links associated to that analytics processor. Then when you generate the source template, it will contain a line of code to inject the associated links into the analytics processor. When you generate the source template you’re also generating underlying structure to access those links.

Reading Data from Your Devices

Use device message links to read data streaming from your devices. See Use Device Message Links to Read Device Data.

Accessing Data Sources

Use links to access different data sources to obtain historical and summarized data to enrich and contextualize the data from your devices. You can then analyze this data to predict trends and future events.

Typically historical data is data from your sensors and devices that you captures before you started using Oracle IoT Cloud Service.

Summarized data is derived from your sensor data. This summary can be the mean value, the standard deviation, the number of occurrences, and similar. You can derive summary data from live sensor data, or from historical sensor data.

You can also use links to write new historical, summarized, or raw data to your data synchs.

You can read data and write data to the following data sources:

• See Use Links to Access Oracle NoSQL Database Cloud.

• Oracle Database Cloud Service

  Typically you Oracle Database Cloud Service to store historical data or summarized data from your Enterprise Resource Planning (ERP) systems. You can use this data to contextualize the date from your sensors and devices.

  See Use Links to Access Oracle Database Cloud Service.

• Oracle Cloud Infrastructure Object Storage Classic

  Typically you use Oracle Cloud Infrastructure Object Storage Classic to upload historical data from your on premises systems. Then you can transfer this data to to improve the time it takes to analyze it using Spark operations.

  See Use Links to Access Oracle Cloud Infrastructure Object Storage Classic.

For more information about the different types of links that you can create, see Task Overview for Creating Links.

Spark SQL

Spark SQL links let you use a SQL query as a data source for your application. The from clause in your Spark SQL link can use any of the links that you configured to use from worksheets.

Spark SQL links also have an option to use them from worksheets. This lets you build a query on top of another query.

You can create a Spark SQL link and then defining the Spark SQL query. See Use Links to Access Your SQL Queries.

If you need to experiment with your data and your query, you can use worksheets and then export your query to a Spark SQL link. See Run Ad Hoc Queries and Reuse Your Queries.

Delivering Results

Use analyzed message links to deliver your result through a data message or an alert.

See Use Analyzed Message Links to Deliver Analytics Results.