Building A Case For Database Migration

Building a Case for Database Migration

Contents

Section Title Page

1.0 Purpose 3

2.0 Why Do Companies Migrate? 3

3.0 What Inhibits a Decision to Migrate? 4

4.0 The Migration Process 5

5.0 Competitive Overview 6

6.0 Summary and Conclusion 6

1.0 Purpose

This document provides a supporting framework for enterprise IT departments to make business decisions on how best to migrate from one database system to another. It is written as a guide to structure a business case for presentation to upper level management.

2.0 Why Do Companies Migrate?

Given the critical nature of database applications in the life of a business, it is often hard to believe that a company would choose to jeopardize its operations by moving to another vendor's database system. Indeed the fact that some business applications were written many years ago and were modified over the years by personnel who may no longer be in the company, it is understandable that there is not only trepidation about making a move to a new database system, but also there may be a high cost associated with the mechanics of the migration. But the fact is that companies vehemently oppose being trapped on one vendor's products because there can be serious consequences such as higher total cost of ownership (TCO), less scalable solutions, a lack of developers, and more. But there are also issues about the flexibility of the business and how it can be negatively affected by the inability to move quickly because of being trapped on a single database platform. The key reasons why companies migrate are:

a) Expensive maintenance contract renewals from a vendor that force a decision to pay or migrate to a less expensive platform.

b) Expensive upgrades by the vendor forcing you to either upgrade or migrate.

c) Pressure to create a consistent face to a company's customers following a corporate merger drives the migration of one company's applications to the database platform of the other company in the merger.

d) Opportunities to leverage On-Demand or Service Oriented Architectures (SOA).

e) Pressure to more rapidly improve or update applications in a dynamic marketplace. One impediment to this is the limited availability of applications development personnel, internally or externally, for some database systems (e.g. Oracle). A natural solution is to migrate to a database vendor that has a strong development environment with many certified developers such as .NET and SQL Server.

f) Pressure to drive TCO down often results in a drive to consolidate platforms, management systems and personnel. A major aspect of this is the migration of applications to fewer server platforms or to less expensive server platforms.

g) Pressure to rapidly add new revenue producing customer service applications usually means that new applications need to be deployed on existing DB infrastructure without need to rewrite the application. In addition, this migration driver relies on the existing DB infrastructure being very scalable and this may drive a company to migrate to a new server platform entirely.

It is very important to understand the key drivers to migrate, because this gives you insight to the form of the value proposition you ultimately use to win approval from executive management for the entire project.

3.0 What Inhibits a Decision to Migrate? What are the key things to consider when pushing a decision to perform a database migration project?

Just as important as it is to understand the drivers for migration, one must understand the inhibitors to migration. Understanding these inhibitors and the degree to which they concern the executive management of the company, is the key to finding the right migration vendor as well as selling your concept up the decision chain. The key inhibitors are:

a) Cost of manual migration is high because of the sheer volume of code requiring conversion, difficult conversion cases due to differences between database systems and the need to perform testing of the converted code to ensure correct conversion. Look for a solution that automates much of the labor intensive work in the migration process allowing the IT staff to focus on the most difficult conversion choices.

b) Time to complete manual migration is another serious issue because companies need to ensure continuous operation, or they need to deploy a new application quickly or they have a need to make a decision on a contract renewal. Look for an automated solution that performs the bulk of the conversion in a matter of hours or days depending on the complexity of the conversion and the number of databases needing conversion. As a result, the time to complete a migration project is primarily the time it takes to perform integration and test and is a small fraction of the time it would take without automation.

c) ROI on the migration project is a question companies have to deal with no matter what the reasons for the migration. But given the need for new hardware, software and the effort to migrate, all coming from different vendors, the ROI can be unclear. Look for a solution that provides an on-demand assessment of the nature of the migration. It must estimate the total manual work required and the level of automation available with the vendor's migration tools on your specific databases. As a result, it would be much easier to perform the required ROI studies.

d) Disruption of business during the migration is a concern. Look for a solution that can be used on a copy of the database. A migrated version can be created, integrated and tested independently of the running database until a cutover is deemed appropriate.

e) Database systems can be radically different because of proprietary extensions to the basic standard database languages (e.g. SQL) and may cause issues in performance or accuracy