Skip navigation
Interior of modern data center with blue lights. Dmitriy Shironosov / Alamy

Technical Debt and the Sustainable Data Center

With data centers being some of the most carbon-intense parts of enterprise operations, CIOs can be the champions of ESG results when they pursue the management of technical debt.

We’ve uncovered another reason for CIOs to loathe technical debt, especially in an era when businesses face increasing pressure to improve energy efficiency: Technical debt wastes energy and undercuts ESG initiatives.

Here's why that's a problem, along with tips on how to make technical debt reduction a part of your data center sustainability strategy.

The energy impact of technical debt

The reason why technical debt – meaning inefficient IT processes or resources that businesses could improve, but which they choose not to improve – harms sustainability is simple enough: Technical debt leads to inefficiency, inefficiency translates to higher resource consumption by workloads and higher resource consumption means that the infrastructure hosting workloads requires more energy to run (not to mention more power to cool).

To put that in a real-world context, consider the technical debt example of a poorly coded application that consumes 20 percent more CPU than it would if its algorithms were optimized. 20 percent higher CPU usage will increase the energy consumption of the application by a similar amount.

Beyond technical debt: Other sources of energy waste

It's not just instances of technical debt narrowly defined that lead to wasted energy. Any resource that you operate in your data center without having a reason to do so leads to higher energy costs.

For example, Kelly Fleming, CIO of Cirrus Nexus, pointed to unnecessary data storage as a source of energy inefficiency. "Failing to control and limit data hoarding can affect not just a company’s bottom line but its sustainability goals as well," he told Data Center Knowledge.

Unnecessary servers or applications have a similar effect. An application instance that developers spin up for testing purposes but forget to shut down, for example, consumes energy toward no good end.

Steps toward improved data center sustainability

Avoiding the types of energy waste described above starts with achieving visibility into which resources you are operating and what the purpose of those resources are.

"Understanding and managing the software and IT landscape" of your business is the key to sustainable operations because it helps ensure that "unused applications will be removed and idle processes stopped," Steffen Wittmann, CTO, LeanIX, told Data Center Knowledge.

He went on to suggest that businesses create an inventory of their IT estate in order to identify what they have running and assess each workload's purpose. "As soon as a basic inventory is created, applications will be mapped to business capabilities and then tied to concrete business values and outcomes," he said. "This is the foundation for app rationalization and for managing risk."

Similarly, Fleming pointed to data tagging as a means of keeping better track of which data exists inside a data center and determining when data should be removed to lower energy consumption. "Data in the cloud often ends up on a variety of orphaned, unused resources of different types, from disks and databases to storage blobs and data lakes," he noted. "Without proper tagging mechanisms in place, it can be challenging to keep track of data resources’ service owners."

Fleming emphasized as well that while there are other methods for improving data center sustainability – such as sourcing power from cleaner sources – that businesses can and should pursue where feasible, simply sunsetting unnecessary data resources is one of the easiest and most effective steps companies can take to reduce energy consumption immediately. It doesn't require complex planning or the acquisition of new hardware systems.

As for simply migrating workloads to the cloud, that's not always an easy way for reducing energy consumption. Cloud-based workloads that are unnecessary, or that are subject to technical debt, will waste energy in the cloud just as they do on-premises, according to Wittmann. "Simply moving an existing software landscape to the cloud by applying the lift-and-shift method will not result in any substantial benefits" from a sustainability or cost perspective, he said.

Sustainability starts with cutting technical debt

Going forward, businesses striving to reduce energy consumption and win on the sustainability front will need to work as hard as possible to fight technical debt and extraneous resources. And although moving to the cloud might seem like an easy way to become sustainable, that's not necessarily the case. 

An unsustainable workload in your private data center will become an unsustainable workload in the public cloud, unless you take steps to optimize it – not to mention determine whether the workload needs to operate in the first place.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish