Article by Steve Biondi
Appropriate mainframe management is a common area of debate among IT professionals. Some operate under the “if it ain’t broke, don’t fix it” mantra, arguing that maintenance and upgrades will keep both the mainframe and the applications that run on it in top working order.
Others argue that the longer you use the mainframe, the more expensive upkeep becomes, contributing to sky high IT expenditures that could be ultimately avoided by doing away with the mainframe entirely. And so the question faced by many IT managers remains: should you opt for a mainframe evolution or a revolution?
A mainframe revolution constitutes a complete switch to another platform in which all applications would be written and run on Linux, Windows or UNIX. The advantage to these platforms is cost efficiency; they’re far cheaper to use.
The problem posed by application rewriting is that the process itself is lengthy, complicated, expensive and can often carry significant risk – such as application downtime and reliability issues. For business critical applications, this type of risk is simply unacceptable and presents a serious roadblock to the possibility of application rewriting.
Alternatively, maintaining the mainframe in its initial state poses its own problems. Mainframe technology can become dated quickly, and as application complexity increases, more space and increased mainframe functionality is required. If a mainframe fails to support advancing technology, application performance is put at stake.
One possible solution to this conundrum is application migration, which involves using modernization tools to transfer existing applications to a new platform. With this approach, a compromise is essentially achieved: the applications are migrated off of the expensive and unwieldy mainframe, but complete application rewriting is not necessary, thus saving significant time and money.
In some cases, this process can cut between three and four years of time off a maintenance project. Moreover, the applications’ integrity remains completely intact with no risk of value loss. In this sense, an “evolution” of the mainframe provides the organization with the benefits of customization with drastically reduced variable costs.
Another possibility is a gradual approach to system development and maintenance on a modern platform. For example, significant amounts of time and money can be saved through the development of a modern PC-based tool, and the quality and versatility of applications can also be improved.
This process can be completed at any stage in the application lifecycle, with the testing environment reproduced by the host so that only the production process stays on the mainframe. In this way, modernization can be a piece-by-piece process that doesn’t tap all existing resources at one given time.
Most likely, undertaking a “mainframe revolution” is likely not the most cost or risk effective approach to mainframe upkeep. Naturally, every mainframe maintenance project has its own unique set of criteria and problem sets, but for many – a less drastic “evolution” approach is likely the more appropriate and sensible option.
Cross-posted from CIOZone