Partitioning Aggregation

Partitioning aggregation processes data in parallel across multiple threads to help increase the performance of aggregation tasks.

  1. Go to Configuration > Settings.
  2. Select the Partition Enabled checkbox.

  3. Enter the statements that should be executed during partitioning in the Partition Statements field. The statements should include all of the rows and each line or statement so it can be preceded in separate threads and/or multiple hosts.

    For example, the statement A-M would be treated as one partition. The IBM Security Verify Access connector would aggregate accounts whose names start with the letters A through M, with A and M inclusive.

  4. Select Save.

Partitioning Examples

The IBM Security Verify Access connector supports the following types of partitioning

Range Base Partitioning

  • Aggregated Account with partitions having range 1-20 – Aggregates all users which start with 1,2,3,…,20.

  • Aggregated Account with partitions having range A-F – Aggregate all users which start with A, B, C, D, E and F.

  • Aggregated Account with partitions having range 1-20 and 30-40 (specify each range on new line) – Aggregates all users which start with 1,2,3,…,20 and 30,31,32,…,40.

Wildcard Base Partitioning

  • Start with wildcard(for example, John*, 1*) – Aggregates all users whose value start with John and 1.

  • Single characters (for example, John, Bill, Mahesh {each name on new line, new line characters are used as a separators}) – Aggregates only John, Bill, and Mahesh.

  • Single numeric values (for example, 100, 12, 46 and so on with each on new line) – Aggregates only users which exactly matches with 100, 12, 46.

  • End with wildcard (for example, *Joe) – Aggregates all users whose value end with Joe.

Note
A new partition can be mentioned in the new line.