Upgrading data center equipment can be expensive and disruptive, so an organization may decide to eke more usable life out of legacy systems and infrastructure. While that may seem like a cost-saving measure, it can cost a lot more in the long run. Outdated, inefficient equipment can rack up technical debt, which will eventually have to be repaid with interest.
The term “technical debt” originated in the early 1990s in software development. Developers who take shortcuts to get software released quickly need to come back and fix the code properly as soon as possible. If they don’t pay back the technical debt, it will cost more, in the long run, to troubleshoot, manage, and modify the software.
The technical debt concept has since been applied to all areas of the IT environment. In the data center, it generally refers to failure to modernize legacy hardware and infrastructure. Maintaining outdated equipment requires more time and effort, reducing the efficiency of the operations team. It can also hinder a data center’s ability to reduce energy costs and increase the risk of unplanned data center outages.
Martin Fowler introduced the Technical Debt Quadrant in 2009 to distinguish between “reckless” and “prudent” technical debt. You should avoid both categories of reckless technical debt because it means you did not know about or ignored the problem.
|
Reckless & Deliberate Deliberately ignoring the best practice without a plan to deal with the consequences later.
|
Prudent & Deliberate Deliberately ignoring the best practice with a premeditated plan for dealing with the consequences. |
|
Reckless & Inadvertent Not following the best practice due to a lack of knowledge without the skills to recognize or rectify the debt accrued.
|
Prudent & Inadvertent Not following the best practice due to lack of knowledge but prepared to learn from and rectify any unforeseen mistakes. |
Every component of the data center environment is subject to technical debt if it’s not properly maintained and kept up-to-date. An IT team may decide to forego a server upgrade because there are plans to upgrade the application that runs on that server in the near term (prudent and deliberate). On the other hand, the team may put off a server upgrade because they don’t have enough resources (reckless and deliberate).
Infrastructure components such as racks and cabinets, cooling systems, and power distribution equipment have a lifecycle just like IT equipment. These components can become outdated and inefficient, resulting in higher energy and maintenance costs and a greater risk of downtime.
Well-run data centers have up-to-date documentation that ensures operational staff knows where systems are located, how they work, and how to resolve problems. Documentation can quickly become outdated without a well-defined process for maintaining it.
Software can show its age pretty quickly. Legacy applications typically lack automated features and require more manual effort to maintain. They often have security vulnerabilities that expose other systems in the data center to cyberattacks. Ultimately, legacy software won’t run on newer hardware or be supported by the latest operating systems.
Technical debt must be managed just like everything else in the data center. The first step is to identify any existing technical debt:
Next, you should develop a plan for paying it off as efficiently as possible. The plan should also include a strategy for minimizing the creation of new technical debt.
Enconnex offers an array of solutions for modernizing your data center infrastructure. A good place to start is by upgrading your server racks to a new, best-in-class solution. Our InfiniRack data center cabinet is configured to order and designed for seamless deployment and easy maintenance, no matter the scale. Get in touch to learn more.