How to Measure Your Cloud Carbon Footprint?

Based on recent research by Bain and Company, only 24% of companies are fully ready to reach sustainability goals. A large part of energy consumption comes from the servers housed by large data centers. The electricity consumption by all of the big five companies (Google, Microsoft, Apple,  Amazon and Facebook) equals the electricity consumption of the entire New Zealand. 

To meet sustainability goals, companies are increasingly focusing on strategies they can implement to use their resources effectively. One strategy is to employ technologies that drive the development of environmentally friendly IT. However, none of these strategies makes sense if organisations do not choose a cloud provider with strong sustainability practices. 

We talked with several experts about how companies can measure their cloud footprint, the best practices for reducing cloud carbon footprint, and the role AI and automation play in minimising cloud energy consumption. 

What Is a Cloud Carbon Footprint?

The amount of greenhouse gases, including methane and dioxide, which mostly come from energy use, waste, and transportation, are released into the atmosphere due to various human activities. The bad impact of these gases is measured by a carbon footprint. 

When it comes to cloud migration, a carbon footprint includes the energy consumed by data centers and the environmental impact of disposing of hardware. Reducing their cloud carbon footprint is imperative for companies to reach sustainability goals by 2050. 

How Can You Measure Your Cloud Carbon Footprint?

Deepak Shukla, CEO of Pearl Lemon, gives constructive advice: “Start by using cloud provider tools like AWS’s Carbon Footprint Tool or Google Cloud’s Carbon Footprint Calculator. These tools offer reports on emissions associated with your cloud usage, helping you track and measure your impact.”

Eve Bai, partnership and operations manager, adds: “You can measure your cloud carbon footprint by tracking the energy consumption of your cloud infrastructure, including servers, data storage, and network usage. Cloud providers like AWS and Google Cloud offer tools and dashboards to monitor energy usage and associated carbon emissions. These tools provide insights into the energy efficiency of your workloads and allow you to calculate your carbon footprint.”

Piyush Mishra, co-founder of CodeCarnage explains that “most major cloud providers already track energy consumption and carbon footprint—like AWS Carbon Footprint Tool, Google’s Carbon Footprint Dashboard. But for an on-premise infrastructure, you will need to gather metrics (like CPU, memory, storage, bandwidth, uptime) regularly. And use monitoring tools (like Nagios, Grafana, or even antivirus suites) to collect them periodically via an exposed REST API.”

Piyush also adds that the “region of your data center or cloud platform also matters a lot, as global energy grids have varying renewable energy mixes, making some grids more carbon efficient than others. Open-source tools like “Cloud Carbon Footprint” can conveniently provide accurate carbon estimates by combining data from multi-cloud and on-premise sources while also accounting for the regional renewable energy mix of the energy grid of the data center.”

What Tools Are Available for Cloud Carbon Footprint Calculation?

From Salesforce Net Zero Cloud to IBM Environmental Intelligence Suite, there are many tools on the market that can help organisations reduce their cloud carbon footprint.

Anandkumar Jeevarathinam from Enable Healthcare Inc explains that “cloud carbon footprints can be measured using tools such as AWS’s Carbon Footprint tool or Microsoft’s Emissions Impact Dashboard. Either of these tools provides insights into the energy consumed by different services across the cloud.”

How Does Cloud Usage Impact Your Carbon Footprint?

Deepak Shukla, CEO of Pearl Lemon, claims that “cloud usage directly affects your carbon footprint through energy consumption in data centers. Storage inefficiencies and constant data transfers increase energy usage, driving up your environmental impact.”

Rafi Friedman, president of Coastal Luxury Nocatee Pool Builders, says: “Broadly speaking, the more we access the cloud, the greater our carbon footprint, but one useful thing to keep in mind here is that cloud-based computers are far more powerful and energy-efficient than anything we could have in-house.”

Best Practices for Reducing Your Cloud Carbon Footprint

Deepak thinks that in order to reduce cloud carbon emissions, companies need to “implement sustainable cloud strategies such as auto-scaling, turning off unused resources, or using multi-cloud setups to leverage providers with lower carbon emissions.”

Paul Raphael Gomes, a digital marketing expert who uses artificial intelligence to help businesses grow online and a founder of BenWil Marketing Agency emphasises two critical ways how you can reduce your cloud carbon footprint:

• Scaling vertically: Set up autoscaling and load balancers so that businesses can react automatically when demand for a certain service decline. By scaling down unnecessary server resources, energy usage can be decreased.

SaaS On Demand – Companies that use an “on-demand” cloud architecture only pay for the API calls they make. By using technologies like Lambda, which is a Function-as-a-Service, one can avoid maintaining virtual computers.

According to Piyush Mishra, these are the key strategies for how companies can reduce cloud carbon footprint:

• Efficient Application Code: For hotpaths in code, programming logic should prioritize lower time and space complexities. Also, memory and compute-efficient programming languages (like Rust) can be preferred in critical sections of code (even if not in the whole code base). Opt for lightweight frameworks to keep your application lightweight. Logic, language, and framework are directly controllable by developers and can significantly undercut cloud carbon footprint.

Server & Cloud Optimisation: Hyperscalers and co-located data centers should be preferred over on-premise setups. Serverless architectures can be used to allocate resources on demand, avoiding idle consumption. Regularly decommission unused or underutilized resources—tools like Google Cloud Recommender can provide optimisation insights regarding such resources. Implement predictive auto-scaling to match resource allocation with demand, preventing over-provisioning during low-traffic periods.

Client-Side Optimisation: Offer dark mode as a default option to reduce client device battery consumption, as your application is also responsible for higher battery consumption in traditional light mode. Implement client-side computations using JavaScript or edge computing to offload work from servers to greener edge computing locations with a higher renewable energy mix in the grid. This reduces the server’s workload, energy consumption, and overall carbon footprint while improving user response times.

How Do Major Cloud Providers Address Carbon Emissions?

Major cloud providers address carbon emissions through various sustainability initiatives in response to growing environmental concerns. Microsoft aims to become carbon-negative by 2030 and plans to achieve 100% renewable energy for its operations by 2025. Google has successfully operated on 100% renewable energy since 2020, focusing on efficiency and a circular economy approach. 

In contrast, Amazon, while the largest cloud provider, does not have a specific target date for its renewable energy commitment and faces scrutiny regarding its carbon footprint, despite its Climate Pledge for net-zero emissions by 2040. Overall, Google leads in sustainability efforts, with Microsoft making significant progress, while Amazon’s commitments remain less clear. These efforts highlight the importance of informed choices for companies seeking greener cloud solutions to drive innovation and environmental responsibility.

Read more about this in the article A Green Cloud Race: Microsoft, Google, and Amazon Compete for a Sustainable Future.

How Can AI and Automation Help Minimise Cloud Energy Consumption?

Anandkumar Jeevarathinam from Enable Healthcare Inc. points out that “Industry 4.0 is allowing green investments as there is a movement towards greener cloud providers and optimisation of how cloud resources and capacities are utilized. AI and automation are also crucial in this case—it is possible to automate the server workloads on off-peak hours to avoid high energy usage and get AI insights to eliminate any inefficient situations.”

Deepak adds that “AI can optimise workloads by analysing patterns and ensuring minimal energy consumption. For instance, AI-powered predictive scaling can reduce unnecessary energy spikes, while automation ensures resources are only used when needed, thus minimising the cloud’s energy drain.”

Eva Bai thinks that “AI and automation can minimise cloud energy consumption by optimising resource allocation and adjusting workloads. AI can predict usage patterns and automate resource scaling, ensuring that energy is used efficiently. Automation tools can turn off underutilised instances, balance loads across servers, and optimise cooling systems in data centers, thereby reducing overall energy consumption.”

Go Green with Deployflow 

Although cloud migration services bring enormous benefits, including scalability cost savings, it can also unlock opportunities like green energy transitions. Businesses must consider their environmental impact. 

By choosing providers who are fully committed to use sustainable products and helping companies on their journey to a greener cloud and optimising resource allocation, companies will be able to mitigate the adverse impact of cloud migration. 

Deployflow is fully committed to helping companies migrate toward a greener cloud and speed up their collective net zero transitions. Explore our cloud migration services to learn how we can help you on your green cloud journey.