A stream definition is created for each type of source of data that contains normalization rule(s) that define the expected specified types, formats, validations (for example, filtering and duplicate detection), error handling, data enhancements, and transformations. It then converts the source data into Gotransverse format that is ready to be processed. Based on the business scenario, records will be mediated and prepared for rating. A separate set of rules is used for the mediation step, which can apply functions such as correlation, aggregation, and record split. While the Mediation Service is able to handle very complex scenarios, there may be outliers that are not handled. Please contact Gotransverse if you have any questions about specific scenarios.

When data is available to be processed, you define a stream run job to process a file of data that matches the data definition.  The stream run can be executed manually or scheduled to run on a recurring basis, which processes available files individually.

During the stream run, designated email recipients will receive notifications at prescribed stages so they are alerted to review job runs and take action, if necessary. Notifications include information such as date, job run name, filename, error reason, and successfully processed count. You can view data records during or after job runs to identify and troubleshoot failures.

Mediation Service processes individual uncompressed usage data files in CSV or XML format. These files are retrieved from an AWS S3 bucket or sFTP website that you supply. Additionally, you can add a manifest file to allow for compressed usage files (BZ2 or Gzip), as well as to perform checksum validation and row counts that ensure the integrity of the file and data before it is processed by the mediation platform.

The Mediation application includes three modules: Run Manager, Tracing, and Setup.

Mediation Run Manager Window

Refer to the following topics for additional information about working with the Mediation application: