Table of Contents

Search

  1. Preface
  2. Introduction to PowerExchange Bulk Data Movement
  3. PowerExchange Listener
  4. Adabas Bulk Data Movement
  5. Datacom Bulk Data Movement
  6. DB2 for i5/OS Bulk Data Movement
  7. DB2 for Linux, UNIX, and Windows Bulk Data Movement
  8. DB2 for z/OS Bulk Data Movement
  9. IDMS Bulk Data Movement
  10. IMS Bulk Data Movement
  11. Microsoft SQL Server Bulk Data Movement
  12. Oracle Bulk Data Movement
  13. Sequential File Bulk Data Movement
  14. VSAM Bulk Data Movement
  15. Writing Data with Fault Tolerance
  16. Monitoring and Tuning Options

Bulk Data Movement Guide

Bulk Data Movement Guide

Writer Partitioning

Writer Partitioning

You can define pass-through partitions at the target instance, or writer, partition point to improve the performance of bulk data movement sessions that have VSAM or sequential file targets. With writer partitioning, a session can use multiple partitions to process SQL inserts and write them to targets concurrently.
Use writer partitioning to improve the performance of the bulk data movement session. Writer partitioning is especially beneficial when a session includes heavy transformation processing.
Optionally, you can also use offload processing to move field-level processing from the target system to the PowerCenter Integration Service machine where the session runs. Field-level processing, such as translating data to the target format, can be resource intensive and increase charge-back costs on a target z/OS system. When this processing is offloaded to the PowerCenter Integration Service machine, the processing runs concurrently across multiple writer partitions. To optimize the efficiency of writing data to the target, use both writer partitioning and target-side offload processing.
Writer partitioning uses pass-through partitions and processes SQL insert operations only. If the partitions receive SQL updates or deletes, the session fails with error message PWXPC_12183. If your data contains many updates or deletes, do not use writer partitioning.
Each writer partition uses a single thread. If you enable multithreading, the session continues to run but ignores the
Worker Threads
setting.
You can specify some session properties at the partition level. Based on the
PWX Partition Strategy
setting and whether you specify the same or different session properties across the partitions, PowerExchange uses a single target connection or multiple connections, one for each partition.
If you use a single connection, a joiner thread merges the data from the partitions before sending the data across the network to the target. The joiner does not maintain the order in which it receives data from the partitions. Consequently, the PowerExchange Listener on the target gets data in an order different from the order in which the partitions processed the data.
You must use a PWX NRDB Batch application connection to the target. Use the
Write Mode
connection attribute to control whether the joiner sends data synchronously or asynchronously to the PowerExchange Listener.
If you want to write data to multiple data sets or files, specify the
File Name Override
session property for the writer partitions. Set this property to point to the appropriate target file for each partition. Contention might occur if multiple partitions write to the same target file at the same time.

0 COMMENTS

We’d like to hear from you!