We’ve had many conversations lately with clients regarding their need to organize, prioritize, or “do something’ with legacy data and sometimes legacy code. The origins of these conversations are varied from moving off a production technology platform, enterprise software migrations, industry regulations, to sometimes corporate bureaucracy demands.
The conversations typically center around having large amounts of seldom looked at data that might be several years old, and having a requirement to hold on to that data for potential historical reporting requirements into the future. It’s the corporate equivalent of having Grandma’s family photo collection. There’s good stuff in there and you don’t want to lose it.
On the cost side, most organizations come to a realization the cost to store the seldom touched data (and maybe the cost to license or maintain the code) has grown excessive. The question comes up – What to do? Lately, we’ve been involved in an increasing number of these conversations.
Every solution is different, but most involve converting the data to a new database platform for easier ad hoc reporting – eliminating the software licensing costs. Other parts of a solution could be moving the data to a less costly hardware platform – cheaper disk or storage solutions. Data storage costs center around quantity and speed. For legacy data, access speed is seldom a consideration, so spend your hardware dollars on the platform requirement that meets the need. And while we are discussing analyzing hardware needs, considering longevity is a big factor here. When you go through the effort and cost to migrate legacy data, you want a time-tested solution that will last for years if not decades.
The most expensive platform you manage is your production server and storage. The data and code that has the least value is your legacy set. The goal is to align the cost of storage and reporting with the perceived value and requirements of the legacy resource. With decades of experience in the typical organization, there comes a time when the massive amounts of data accumulated needs to be managed differently than what’s currently considered the production set. That’s where these conversations start.
If you’d like to discuss your options regarding managing legacy data and code, I’d love to talk. Feel free to contact me