Stream Processing¶
Stream processing involves making changes to streaming data to generate a required output. ESB Streaming Integrator allows you to carry out a wide range of operations to process streaming data. These operations are supported via Siddhi Extensions. These operations can be broadly categorized into five categories as follows:
-
Filtering only the required information for further processing out of all the extracted/received data, and other cleaning activities such as pre-processing, adding missing values, removing unnecessary attributes, etc.
-
Transforming data from one format to another, performing a range of mathematical, regex, string, unit conversion, map, json, etc., and enabling the creation of scripts that perform transformations.
-
Joining streams of data with each other as well as static data (such as tables) to enrich the data
-
Performing long-term time based summarizations, as well as performing summarizations for subsets of data based on time or length (number of events e.g., every 10 events).
-
Identifying trends and patterns.