Support and Upgrade Process Deviations

Contents

• Introduction
  • Deviations from standard support and upgrade process
  • Reasons why the Patch and Upgrade Processes aren't following the standard processes

Introduction

This section called 'Analysis Models - Support and Upgrade Process' gives information and guidelines related to the support and upgrade process of the Analysis Models.

Note: The Analysis Models were formerly known as IFS BI Analysis Package.

Note: Related example content in the Analysis Models and IFS BI Analysis Package, such as SSRS and Power BI reports, are not supported.

Deviations from standard support and upgrade process

The Analysis Models do not fully follow the standard support and upgrade process.
Exceptions to the process:

• Single patch deployment must be handled manually.
• Upgrades are not supported, no upgrade scripts for instance.
• Patches do not comply with Delta Engine syntax/parsing requirements.
• Normal build and delivery processes do not fully apply.
  

Reasons why the Patch and Upgrade Processes aren't following the standard processes

Installing and especially updating a Data Warehouse is not as straight forward as performing the same operations for IFS Applications. Some reasons for this are:

• For many of the involved files it is rather difficult to automatically, but also manually, support merge of corrections between versions of files.
• If files need to be modified, it is not as easy as solely deploying or executing after modification.
  Moving information from a source system to a target database means moving large amounts of data, so the ETL process must be carefully planned, both the initial load as well as how to handle customizations and bug patches. Since IFS does not support single bug patch deployment, a decision must be made how to handle the data in the ETL process. Some options are:
  • Patching an existing instance.
    This will probably be the most common scenario. Each modification has to be analyzed and manual corrections must be made, also considering customizations. It is also important to carefully plan how to deploy the corrections.
    Some examples: Should tables be completely regenerated or should only new columns be added? Is it necessary to update new columns in the Data Mart tables? How to update and test SSIS package modifications etc
  • Complete rebuild of the BI Instance.
    This scenario is not likely to happen very often since it means that all already transferred data will be reprocessed.
    Note that if rebuild of an existing instance is planned, it must be made sure that databases/objects etc that shall be replaced are removed before re-execution.
    Upgrade scripts are not provided in support corrections.
  • Creating a completely new instance.
    Instead of rebuilding an existing instance, a better way would be to create a totally new BI instance. This is a rather straightforward scenario. The problem is still that it might take a long time to initially load the ETL part of the new instance.