Table Parameterized Functions

A Table parameterized function (TPF) is an extension of a table UDF that accepts table input parameters in addition to scalar input parameters.

You can configure user-specified partitioning for your TPF. The UDF developer can declare a partitioning scheme that breaks down the dataset into smaller pieces of query processing that you distribute across multiplex nodes. This enables you to execute the TPF in parallel in a distributed server environment over partitions of row sets. The query engine supports massive parallelization of TPF processing.

• Learning Roadmap for TPF Developers
  Develop a C or C++ TPF.
• Developing a TPF
  Review the major steps required to develop a TPF.
• Consume Table Parameters
  A table parameter is a non-constant parameter. This means that the TPF must be in the execution state to retrieve table parameters.
• Order Input Table Data
  Either a SQL Analyst or the UDF developer can order incoming data.
• Partitioning Input Data
  Use the PARTITION BY clause to express and declare invocation partitioning in your parallel TPF.
• TPF Implementation Examples
  Implementation examples start with a simple TPF and increase in complexity and functionality as the examples progress.