Aberdeen Group - Home
Facing the Challenges of Modernizing Mainframe Data
Ed Franklin
MAR 04, 2016 18:51 PM
A+ A A-

Facing the Challenges of Modernizing Mainframe Data

Network servers in datacenter

An IT manager recently told me he hadn’t touched the applications on his mainframe since Y2K. For many other organizations, it has been even longer than that — and for good reason.

Over the last 25 years, mainframes have been a solid investment in term of reliability, performance, and security. It’s estimated that within the insurance industry, half of all core applications are still on a mainframe. But for insurers and other data-intensive industries that rely on mainframe applications and data, those days are coming to an end.

The modernization of mainframe applications is now a business imperative.

Guest article by Ed Franklin, VP of Global Marketing, TmaxSoft

That can be a scary proposition for those that have relied on their mainframes for decades. But mainframe apps are simply not compatible with the fundamental drivers of today’s business and technology paradigm.

The main drivers for mainframe modernization

To begin with, mainframes are running COBOL, PL/1, or other languages difficult to support. They cannot be easily expanded or quickly adapted to serve new market opportunities. In fact, many users may employ mainframe software and tools whose vendors may no longer exist, so maintenance can be an issue as well. Mainframes also demand an unreasonable amount of resources in terms of power and cooling costs.

Big Data and analytics are driving innovation and competitive differentiation for organizations across all industries. But how do you unlock the tremendous value hidden with mainframe data? The legacy data structure in mainframes often doesn’t play with analytics. But organizations will have to make their mainframe data compatible with analytics in order to develop innovative new products and services, take advantage of new market opportunities, and make themselves stand out from the competition.

Then there’s the cloud. Many companies move to cloud for increased capacity and higher ROI from their commodity x86 systems. These advantages are significant reasons for modernizing mainframe applications as well, by separating database and application tiers and employing modern SQL-based databases. The result is a legacy application with better scaling, higher reliability, and improved integration with big data analytics.

It’s also true that if you want to compete and succeed in today’s marketplace, you must have the ability to conduct commerce on mobile apps. But mainframe apps do not interface with mobile apps. Once again, organizations will be pressed to modernize mainframe data so it can be used with efficient, cost-effective x86 hardware, and multi-tier cloud architectures that serve up valuable mobile apps and data to be consumed by customers, partners, and employees.

But modernizing is not as easy as it seems 

With this urgency to modernize mainframe applications, organizations are faced with some tough choices. Attempting to rewrite the entire database and messing with the business logic can be very risky. You may need a whole programming team and the budget hit will be significant. And because many mainframes have been in operation for a quarter of a century, it can be uncertain how re-writing will affect your downstream processes.

Organizations must realize that the process of decommissioning mainframes is just as important as the introduction of new systems. Smart planning cycles and formalized processes are required.

The good news is that there are now options for modernizing mainframe data without a complete code rewrite. Automated modernization and rehosting solutions can bring pre-SQL, single-tier mainframe applications to SQL datbase in a multitier, x86 cloud environment — without any change to applications.

Transitions can be largely automated, and testing tools can identify potential downstream glitches. Implementation can become the final modernization strategy or can be used as a safe cloud-based operating stage for a complete local re-architecting. Payback is typically achieved in 12 to 18 months, unlike more expensive alternatives.

Once the constraints of the mainframe have been lifted, all of that valuable data can easily make its way to cloud, mobile, and analytics platforms. And from a business perspective, that’s when the most rewarding benefits of modernization can begin.


[%= name %]
[%= createDate %]
[%= comment %]
Share this:
Please login to enter a comment: