DEVOPSdigest

When we think about data, we view it as a concept, not something with physical form. Despite its intangible quality, the massive data flow significantly impacts the environment. Data centers and transmission networks emit nearly 1% of energy-related GHG emissions, and mid-sized centers use 300,000 gallons of water(link is external) every day. With data and server traffic growing exponentially, those numbers will likely grow while energy prices skyrocket. Additionally, the government is considering environmental, social and governance (ESG) reporting requirements(link is external), which would force companies to be cognizant of their byproducts. As companies focus more on both cutting costs and ESG efforts, IT leaders will find themselves shouldering more responsibility for energy consumption and the resulting environmental footprint.

Why IT leaders?

Because they own many moving organizational functions that contribute to energy usage. Strategies to measure IT's impact are still being developed, but now is the time to start considering strategies to demonstrate reduced energy use. Software developers and DevOps teams can contribute to the effort. The concept may seem abstract, so let's take a closer look at ways DevOps can impact energy costs and carbon footprint.

Creating More Efficient Applications

Developers should work to write more efficient code — code that doesn't strain CPUs by running continuously and requires fewer infrastructure resources to execute.

Minimizing monolithic app creation promotes better environmental practices. A monolithic app contains most or all of its functionality within a single process or container. When scaling the app, the code will be deployed several times. If you only need to scale one component, you still must retest the entire application, stop and start service and redeploy all the code each time, generating significant CPU usage.

The monolithic approach is becoming obsolete with the cloud infrastructure. Developers should be conscientious as they migrate legacy software to the cloud. Rather than just transferring the entire monolithic application, consider refactoring the code to be more cloud native and efficient. While developers don't make the decisions about the move to a hybrid cloud, they can have a positive impact by being more aggressive in streamlining code.

The cloud and virtualization will help improve CPU utilization. Hyperscale cloud providers are significantly more efficient than data centers because of their economies of scale. They can be more economical in their HVAC systems, lighting and water usage, which drives down overall energy consumption in ways that individual companies cannot duplicate.

Implementing Observability

Performance tuning requires observability implementation. IT leaders must develop a way to monitor total infrastructure and CPU usage. Once teams collect application performance data, the engineering organization needs to action it, resulting in a code change.

Many applications have different performance characteristics in production than in test environments. As such, frequently deploying performance optimizations generates the data needed for the next round of improvements. The observability tools can also implement automated canary analysis and automatically verify that the updates are having their expected positive impact.

Continuous deployment provides the infrastructure to allow companies to rapidly iterate on and fine-tune these performance optimizations. Teams can measure deployment effectiveness through DORA metrics, standard DevOps measurements encompassing:

■ Deployment frequency

■ Lead time for changes

■ Change failure rate

■ Mean time to recovery

By improving upon these numbers, teams can more efficiently optimize applications to require fewer resources.

Aligning ESG with Business Needs

In the end, a company's priority is revenue. With rising energy costs, changing consumer preferences and expanded reporting, ESG is finding more alignment with corporate goals.

Efficient programs improve customer experience while reducing energy consumption and spend, benefiting the company's bottom line and the environment. Businesses can market ESG. Many customers prefer greener companies, so being perceived as driving energy savings gives businesses a competitive edge. With the looming possibility of government-required ESG reporting, companies have more incentive to monitor their activities and quantify their efforts.

As ESG initiatives grow in importance and alignment with business priorities, company leaders will focus more intently on decreasing their environmental impact. While IT departments, specifically developers and DevOps teams, are not solely responsible for reducing the carbon footprint, they will be expected to contribute. Sustainability should be a priority in software development from the first line of code, and IT leaders need to consider how to use metrics to demonstrate their ESG efforts if and when reporting becomes compulsory.