Make vs Buy – The Data Movement Dilemma

Published in TDAN.com January 2000

Although solutions are available to manage and automate data movement across heterogeneous systems, research indicates that most companies still spend far more on developing their own scripts and
utilities to move data than they do on third-party solutions.

However, recent evidence suggests third-party data movement products provide an advantage over in-house developed solutions for an overwhelming number of business enterprises. Given that more and
more companies have to apply their inside development resources to the projects that need them the most, it is critical to ensure that the major issues are taken into consideration during the
make-buy evaluation.

THE CASE FOR MAKE:

A valid argument can still be made for developing in-house data movement software. For example, there may be some instances where an organization’s needs are so unique as to require custom
developed data replication software.

In other cases, the data movement needs may be so rudimentary (e.g. basic replication of minor amounts of data between two identical database environments in a localized department) that it is
difficult to justify the expense of licensing a solution. If these scenarios describe a company’s requirements and it has the capability to develop an in-house solution without affecting other
customer deliverables (as well as absorb the ongoing maintenance and upgrade of the solution), then a make decision is most likely the proper one.

THE CASE FOR BUY:

In many corporations the benefits of licensing third-party software to meet enterprise needs are quite compelling. Data movement software benefits companies by:

 

  • Lowering lifetime costs
  • Taking advantage of the vendor’s support, maintenance and upgrade capabilities
  • Increasing adaptability
  • Allocating internal resources to meet other unique organizational needs
  • Shortening implementation time, thus accelerating return on investment
  • Leveraging the vendor’s investment in research, development, skills and continuous enhancement of its products

 

In most cases data is moved between different databases or database schemas, or involves significant amounts of data that needs to be moved among several locations. Available third-party
alternatives offer data movement for most popular hardware, operating systems and database management system platforms. Some also contain sophisticated data transformation capabilities that can
handle just about any requirement.

ISSUES TO CONSIDER:

Environment Characteristics: This is the one area that companies frequently overlook prior to embarking on a data-movement development project. It is also an area that can have tremendous impact in
the complexity and cost of developing the data movement software. Are the databases different (type, version, schema)? How frequently does the data change? How often should the data be moved? Does
the data need to be transformed in any way? How well will the network and CPUs support the additional load of the data movement process? This is an abbreviated list of the types of environmental
issues that will impact the data movement process.

LIFETIME COSTS: When comparing costs of developing versus buying data movement software, it is important to consider total lifetime costs. While the cost of a new development can be
substantial, it is frequently only a fraction of the lifetime costs involved in maintenance and support. The company that develops software must, on an ongoing basis, provide support, fix bugs, add
new features and adapt to new hardware, software and application platforms.

FLEXIBILITY: In many cases, in-house developed data movement software is developed for a specific business information requirement. However, as business requirements change, it often
causes a significant effort and expenditure in development, testing and migration to support the new requirement with in-house developed software. In most cases, third-party software is developed
with a flexible architecture that allows the business to easily modify its data movement processes to meet changing needs.

OPPORTUNITY COSTS: Another factor not often properly considered is opportunity costs. When comparing licensing software vs. developing it in-house, an organization must decide how to price
in-house resources (salary, overhead burden, capital equipment). Where an organization must allocate IT resources from a pre-set budget, using a resource on one development project means that
resource cannot be used on another project. Deciding to develop data movement capabilities in-house means that organizations must, at best, defer other applications or, at worst, not undertake them
at all. The foregone benefit of the project not chosen should be factored into the cost of the data movement development project.

The bottom line in weighing in-house vs. licensed alternatives may be that most organizations today have sophisticated longer-term data movement needs that can’t be met with homegrown solutions.
With the proliferation of heterogeneous database environments in a mix of legacy, client/server and Web environments, few corporations have the in-house expertise to develop sophisticated data
movement solutions. Management is beginning to feel the pain and is finding ways to relieve it.

After considering the issues above, it is easy to see why the major market research firms are expecting a dramatic increase in the number of companies that adopt third-party software for data
movement.

Share

submit to reddit

About Doug Prouty

Lakeview Technology

Top