How Do We Do This
Aiming to do things more green is not always easy. Like with any other aspect of software development, there’s a question, “Should we do X?” which is usually followed by the answer, “It depends.” and “How do we measure it?” Now, we still need many things from the industry, like cloud providers, to succeed in our journey. Like any great thing, this will also require a lot of cooperation.
But just because something is hard doesn't mean it’s not worth doing. In our case, worth is everything; it’s our environment.
In the following chapters, we’ll discuss how to bring a sustainability dimension into software development.
Collaboration and cooperation
First on the list and for a reason.
In order to amplify our message, we need everyone’s help. No matter your industry or how big or small it is, we need our shared voice. Many of us are now realizing that there are also other kinds of value to be created, beyond, for example, just profit.
However, the biggest realization is that sustainability and profitability are not mutually exclusive. So, for anyone reading this, please reach out to us, and we will do our best to help you within the limits of our abilities.
Now that we have been heard, it's time to focus on the infrastructure that enables the whole digital world.
IT infrastructure provider's responsibility
Let's take the point of view of an IT professional. We believe that the biggest impact can be made at a macro level: any optimization done at this level will benefit all the customers using the cloud. Cloud providers are already doing good things regarding sustainability, like multi-tenancy and aiming to use carbon-free energy sources, but there’s still a lot to do. For example, data centers use vast amounts of fresh water for cooling, which erodes the environment.
Our part is to insist that cloud providers be more transparent about their impact on the environment and demand better tools for deciding which cloud provider or region to use at what moment. For example, we need clear cross-cloud comparable Carbon dioxide equivalent (CO2e) meters from all the cloud providers. Think of it as the cloud cost now being shown by the minute or second.
Regarding CO₂e meters: Regulations such as the Corporate Sustainability Reporting Directive (CSRD) driven by the European Commission are coming. Through subcontracting chains, CSRD will affect all companies. These regulations help us to make more informed decisions. This leads to minimizing our environmental impact and promoting the best sustainability champion. CSRD will generate more work for organizations, but we want to be there to help when the time comes.
Next, we’ll tell you what we do at a micro level.
Hands-on Green Lean
The Green Lean is a value optimization practice that combines disciplines and approaches like FinOps, performance engineering, GreenOps, and green code. Our goal is to save natural resources by reducing cloud costs and improving cloud efficiency.
First, we start at the infrastructure level with FinOps and GreenOps. The most optimal outcome happens at the intersection of these two. A classic example would be using smaller or fewer virtual machines to complete the same task as before. An even better outcome comes when we eliminate waste by removing resources (or whole services) that provide no value.
Sometimes, the root cause of inefficiency and waste lies at the software level. In that case, performance engineering is needed to solve the problem: Optimizing code for less resource utilization. Again, the most optimal outcome would be at the intersection with green code principles, but the relation between these two is not as straightforward as at the infrastructure level.
Here are some of the ways our services can help at various stages of your journey in the cloud:
We evaluate the most efficient way to host their workloads in the cloud, taking into account both the software and cloud architecture.
We accelerate the adoption of cost and resource efficient cloud services for running new workloads.
We improve the visibility on cloud resource cost and efficiency, integrate cost tracking as part of development practices, and identify and optimize the factors that contribute the most to the cloud bill and carbon footprint.
We guide the way towards hybrid cloud solutions to adapt to regulatory and financial restrictions.
“Cloud service providers sell you on the idea that you pay for what you use. However, a more accurate statement is that you pay for the resources you provision, whether you use them or not.”
Showing new ways to look at things
Two of the most significant environmental impacts come from IT infrastructure energy consumption and consumer electronics (e-waste). Both are steadily rising as the digital world keeps growing. We too, love our gadgets and streaming services; after all, we are interested in all things technical. But this does not mean we can’t be less excessive with those things. Reducing the consumption of electronics is the most effective way to help our environment, but there’s still more to do.
One way to reduce e-waste churn is to extend the life cycle of old devices. Device manufacturers and app developers can do this by supporting their older devices longer. If the hardware boundaries of old devices limit the software, software developers can use strategies such as progressive enhancements to ensure all devices can enjoy the product at some level. This reduces the novelty of owning new devices, but perhaps that novelty can be achieved in other ways.
We should also reconsider whether everything today needs to be designed for high availability. By default, cloud best practices encourage us to design systems to be available across multiple data centers, edge locations, and perhaps even multiple geographical locations. However, we often forget to consider whether all of it is necessary for the software services we are delivering. Reducing redundancy not only lowers costs but also simplifies architectural patterns.
Lastly, does the world need every piece of AI-generated content? All this newly generated data—which uses expensive computing capacity—must be moved and stored somewhere, using resources. It has become too “easy” to generate any content—which no one uses—with a click of a button. AI can be used for good, but it's important to contemplate whether we need this content or not.
As stated earlier, infrastructure and service providers have the biggest responsibility. The aim is not to make our lives harder but to help us see things from new perspectives. Going deeper here would make the text too long, but for now: we want you to join us on this learning path.
Being hardware efficient means ensuring that every hardware device is utilized as much as possible for as long as possible. Learn more from the free course created by the Green Software Foundation and provided by the Linux Foundation.
Consider the tradeoffs
We must recognize that many of our infrastructure or software decisions involve tradeoffs. In FinOps, maximizing cloud value often means using more capacity, like computing power. Similarly, in performance engineering, optimizing code may reach a point where further performance gains require different approaches, like increased hardware capacity.
Consider the example of using older-generation computing hardware. While they may be less energy-efficient, they can be cost-effective and environmentally beneficial when fully utilized: The energy that went into producing the hardware is compensated further by utilizing the hardware. Eventually, though, performance demands may necessitate upgrading to more efficient hardware.
About our webpage: A simple, text-only HTML page would consume less energy, but we use richer content to achieve our goals better. That said, small optimizations across systems can accumulate into significant improvements. The key lies in a holistic approach, where knowledge sharing and culture drive change. Ultimately, companies decide how to manage their saved resources in a free market.
A new kind of value
There is a phenomenon that as technological improvements increase the efficiency of resource use, the overall consumption of that resource may actually rise rather than fall. This happens because increased efficiency lowers the cost of using the resource, which can lead to greater demand and, ultimately, higher total consumption. This is called the Jevons Paradox. To counter its unintended consequences, CSRD offers one approach, but broader, more combined efforts are needed to address the issue fully.
We believe that our customers understand that there are other types of value returns than pure money: today's consumers also value quality and durability. People expect more and more responsibility from companies. We agree. People deserve more!
Supporting environmental organizations economically and transparently is one potential path. This is our path, and we will take it: 10% of our profits will go back to nature.
Nature is the thing that truly matters.
Read here what is our why.
