You can test your StreamBase application by running it, then sending it a collection of test data in the exact format expected by your application's input streams. A collection of such test data is called a feed simulation.
Feed simulations have the following characteristics:
-
Feed data can be generated automatically from random data, each field according to its expected data type.
-
You can customize your feed data to contain actual sample data, or realistic data of the exact format expected by your input streams.
-
Your feed data can be formed from a mix of automatically generated and customized data.
-
You can have feed data read from a comma-separated or tab-delimited file, from a table in a JDBC database.
-
You can provide Java code for a custom file reader for non-standard, proprietary, or binary files to serve as the source of feed data.
-
You can ramp up the data rate of your simulated feed to stress-test your application.
-
You can save feed simulations as disk files to be shared among developers or used in different projects.
-
You can edit saved feed simulation files.
You can run feed simulations in several ways:
-
In StreamBase Studio, you can send data manually, one tuple at a time.
-
In Studio and on the command line, you can run a generic feed simulation that generates random data.
-
In Studio and on the command line, you can run a saved feed simulation file.
The following topics provide the details of creating, maintaining, and running feed simulations for StreamBase applications.
-
See Running Feed Simulations to learn how to run saved simulations in StreamBase Studio's Feed Simulation view.
-
See Manual Input of Data to learn how to send data to a running application one tuple at a time.
-
See Using the Feed Simulation Editor to learn how to define and save feed simulation files.
-
See Feed Simulation with a JDBC Data Source to learn the details of using a SQL query to a JDBC database as the source of input tuples for a feed simulation.
-
See Feed Simulation Timestamp Options to learn the details of using the Timestamp from column and Include in synchronized timestamp group options of the Feed Simulation Editor.
-
See Feed Simulation with Custom File Reader to learn about writing and specifying custom Java code for reading non-standard, proprietary, or binary files as the source of a stream of input tuples for feed simulations.
-
See Map to Sub-Fields Option to learn about mapping hierarchical data in CSV files.
-
See Command Line Feed Simulations to learn about the command-line utility that runs saved feed simulations at the Windows or UNIX command prompt.