Db2

 View Only

Best Practices for Building a Data Migration Strategy with IBM InfoSphere Optim High Performance Unload

By Arun Ramachandran posted Thu December 01, 2022 12:41 AM

  

Executive Summary

This paper targets people involved in planning, configuring, designing, implementing, or administering a data warehouse that is based on DB2® Database for Linux®, UNIX, and Windows® software. The examples in this paper apply generally but are focused on the IBM ® PureData System for Operational Analytics.

IBM InfoSphere Optim High Performance Unload for DB2 for Linux, UNIX, and Windows V4.02 is a high-speed tool for unloading, extracting, and migrating data in DB2 for Linux, UNIX, and Windows databases. High Performance Unload (HPU) is designed to extract and migrate data from DB2 table space containers. A data migration strategy that uses HPU minimizes storage needs and automates many manual tasks.

HPU uses named-pipes and parallel LOAD operations to stream data from the source to the target database, minimizing the need to stage the data on disk. You can direct HPU to determine where different partition maps, software levels, and key constraints are used and automatically handle these events during data migration

HPU can also unload subsets of data from target database without the need to access the DB2 software layer. With this functionality, you can migrate data online from larger production systems to smaller pre-production or development environments.

HPU has other uses in data extraction from DB2 backup images which are covered in the companion paper Best Practices Using IBM Optim High Performance Unload as part of a Recovery Strategy in an IBM Smart Analytics System.

Introduction

This paper describes best practices for incorporating the use of HPU into your migration strategy and implementing HPU for an IBM PureData System for Operational Analytics data warehouse. This paper covers how to migrate different sets of data between two databases.
To use this paper, you should have a basic knowledge of HPU software as well as DB2 software as implemented in a partitioned data warehouse environment. The further reading section in this paper contains links to product documentation and papers that are referenced in the paper.

The first and second sections outline the benefits of using HPU and the possible considerations of integrating it into your system.

The third section reviews the HPU Migration Process, HPU control files, HPU parameters and recommends best practices for creating control files.

The fourth section provides details on how to install and configure HPU for migration in an IBM PureData System for Operational Analytics.

The fifth section offers specific best practice recommendations for different scenarios.

The appendix section outlines the configuration of the test system that was used in developing this paper.

Download the full report for more on building a data migration strategy with IBM InfoSphere Optim High Performance Unload.

Download the report to get started! 

This paper builds on and complements a series of best practices papers that discuss aspects of a DB2 data warehouse. Refer to the Further reading section of the full report for more information.



#Db2
0 comments
15 views

Permalink