Migrating from legacy applications such as content management or business intelligence systems can be an intimidating and expensive task for many companies, but those who choose to undertake such projects often reap long-term benefits.
“It’s kind of like going to the dentist for a root canal,” Mike Zucker, the University of Miami’s IT applications development director, says. “It’s not terribly enjoyable at the time, but if you know it’s necessary, it’s worth it.”
The University of Miami had 42 integrated database management system (IDMS) applications, the majority of which were developed in-house, running human resources, payroll, general ledger, and other student systems. Zucker said that the homegrown systems had been continually muddled together over the years, creating a daunting migration task.
“And we had been doing something that’s been in database books since the late 70s and early 80s, which was integrating everything,” Zucker said. “This seemed like a good idea at the time, but in the end, we we’re left with one giant application which was all in IDMS. So, you can start to get the picture of this horror show.”
With over nine million lines of code connecting the school’s payroll system to its human resources system, Zucker said he knew the school had to upgrade before the situation got any worse. For enterprises in the same situation, weighing options such as the type of migration to make, what it will cost, and whether it’s even necessary are essential steps before going through with the project.
Upgrading versus replacing
Many companies are deciding to upgrade rather than rip and replace their old systems. According to Kevin Quinn, vice-president of product marketing at New York-based Information Builders, if companies are using applications that still have value and contain lots of significant data, “ripping and replacing really shouldn’t be an option.”
At the University of Miami, Zucker felt it was important to upgrade the homegrown systems, as opposed to replacing them with new products, in order to truly cater to the needs of students. He said the applications were something that he could continue to evolve in an updated infrastructure.
“To buy an off-the-shelf product, we would end up looking like every other university, and that seemed like a big disadvantage for us,” Zucker said.
Dallas, Tx.-based Ateras offered an automated solution for upgrading the code, without putting the systems out of commission and offline. The theme of the conversion, as proposed by Ateras, was that “we hope nobody notices the change.” This is something that can only be achieved via upgrading as opposed to starting from scratch. Ateras said the university migrated to a DB2 CICS COBOL environment at the cost of US$2.5 million.
In making a decision, Zucker weighted that cost versus with the licensing fees the school was paying for the IDMS system and decided the conversion would actually save the school money.
“These are our day-to-day, mission-critical systems that are constantly evolving while running against an obsolete database,” Zucker said. “Retaining programmers, attracting new programs and using new tools and methodologies was pretty much impossible.”
Evaluating the cost and time versus the reward
As seen in the Miami example, the return on investment is a crucial aspect to evaluate after deciding on an upgrade. Andy Aicklen, vice-president for HP Canada software, said that companies who have limited resources have to make sure the upgrade has value and they are not simply dumping money “into an empty hole.” Even companies with lots of resources should avoid doing this, Aicklen said.
Calgary-based Good Earth Caf