Developing edge applications with IBM Edge Analytics

You can develop streaming applications that are run on an edge system or an IBM® Streams service instance in IBM Cloud Pak for Data.

Before you begin

You must install the Streams service on Cloud Pak for Data. The Streams instance is used to develop and build the applications that you run on the edge systems. For more information, see Installing Streams.

After the Streams service is installed, you must provision a Streams service instance in Cloud Pak for Data. After you provision a Streams service instance, you get a Streams instance and a build service, ready to build, deploy, and monitor your streaming applications. For more information, see Provisioning Streams.

About this task

Limitations:

The Import and Export operators are not supported in edge applications.
The Pub and Sub operators are not supported in edge applications.
In Streams Flows, the Streams operator (source and target) is not supported.
Consistent regions are not supported in edge applications.
All edge applications run as a single processing element.
Edge applications must be built on the Streams service in Cloud Pak for Data. Streams Studio is not supported for developing edge applications.

For more information about using Streams, see the IBM Streams in IBM Cloud Pak for Data videos.

Procedure

Choose the appropriate development environment for your edge application:

Streams Python API
Use this API to develop streaming applications that will run on the Streams instance in Cloud Pak for Data.
- Python API in a Cloud Pak for Data notebook.
  You must install Watson™ Studio before you can use this option. For more information, see Installing Watson Studio.
- Python API without a Cloud Pak for Data project.
For a list of supported Python toolkit packages, see Supported Python toolkit packages for Streams.
Streams Processing Language (SPL) in Visual Studio Code:

SPL applications can use the full set of Streams capabilities. You can use Microsoft Visual Studio Code as your editor.
Streams Flows from within a Cloud Pak for Data project:

Streams is a drag-and-drop editor for constructing simple Streams applications, so it is a great way to get up and running quickly. You must install the Streams Flows service to use this option. For more information about installing Streams Flows, see Installing Streams Flows.

What to do next

After you develop an edge application, you can use an IBM Streams development environment or a Streams Python API program to build it. For more information, see Building an edge application.