Page tree

 

The following diagram presents a high-level overview of the components and services in the Trifacta® platform.

Figure: Platform Components and Services

  • User Interface: Depending on the product, users interact through one of the available interfaces through a secure connection with the platform. 
  • Trifacta Services: The following services interact with flow and dataset metadata in the Trifacta database, a proprietary database.
    • Compiler: The Wrangle compiler renders the steps created or selected by the user into a sequence of commands that can be executed against the specified target. See Wrangle Execution Layers.
    • UDF: As needed, developers can create user-defined functions (UDFs), which can be made available to end-users through the UI. For more information, see User-Defined Functions.
    • Scheduling: The Scheduling service enables management of asynchronous execution of jobs on a cron-based schedule. For more information, see Overview of Scheduling.
    • Machine Learning: The executed Wrangle recipes are analyzed for patterns, which provide more focused recommendations to end-users in the environment. 
    • Metadata Server: This server manages the read/write operations of the Trifacta database.
  • Workload Management: This layer, which includes job execution and monitoring functions, is managed by Batch Job Runner, a proprietary service of the Trifacta platform
  • Execution Environment: The Trifacta platform  can execute Wrangle recipes through a variety of running environments.
  • Storage: The unified storage layer provides brokerage services for read/write operations against HDFS, Hive, S3, and other clustered storage.

This page has no comments.