Boston – IT shops have little choice but to modernize mainframes as monumentally large databases hinder agility...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
in development and mainframers retire.
But the planning process for a mainframe or multidatabase migration is daunting, especially for companies with huge amounts of data or many scattered databases. Emily Brand, services delivery manager at Red Hat, offered solutions to the mainframe modernization dilemma during a session at Red Hat Summit 2013 here this week.
Brand suggested that companies start by identifying the applications you need to migrate, then integrate them using an intermediary layer. She suggested Enterprise Data Services (EDS) for this, but other database virtualization software will do. Once the virtualization layer is in place, perform regression and compatibility tests, connect the apps, and port away.
The expectation here is that once the databases are migrated and modernized, your aging mainframe behemoth will be put out to pasture, but whether you retire the mainframe or dedicate it to other tasks, the end result will be a more agile, unified database. And that, Brand said, is part of the ideal environment.
Standardization and consolidation are important to a healthy and future-proof environment, according to Brand. With fragmented databases and custom code for each separate database to be able to communicate with each other and with end users, productivity can be stifled.
An attendee of the talk, Kevin McGibney, services delivery manager for TechFlow Inc., a San Diego-based IT services company, agreed with her assessment. What Brand showed -- a fragmented IT environment running multiple databases that all need to speak to each other somehow -- is the environment of typical database users.
One of the benefits to consolidating and standardizing databases is that it allows developers to try new things with the data. In Brand's example, a user wants the data in a Ruby-based application rather than the company's typical Java. By installing an extra layer -- such as EDS -- just above the databases, you connect the databases under a common code instead of an individual, fragmented code written by different departments.
Since this method decouples business applications from databases, it should make transitioning projects to new teams easier, especially if you standardize and update your programming languages at the same time, Brand said.
Consolidating technologies also means teams don't need so many skill sets, and it makes building teams for every project easier, McGibney said. In other words, there's no longer a need to create separate teams for MySQL, COBOL or other applications.
Regardless of whether your team needs to be more agile or your databases do, Brand said, modernization is the way to go, and now is the time.