What is sustainable coding? (Emerging Tech Series: Part 6)

Key takeaways:

• With increased pressure on businesses to become more sustainable, coding shouldn’t be ignored
• Adopting sustainable coding can help businesses reduce their carbon footprint
• Six ways to improve sustainability through coding practices

Sustainability has been a major buzzword of the last few years, and in many ways, it has not yet fully made the rounds within the software engineering and data science world. Still, the importance of pursuing sustainability holistically across businesses has never been higher.

While there’s broad support for reversing the conditions that have led us to the current climate crisis, many businesses face considerable challenges in reducing their carbon footprint and developing more sustainable ways of doing business. Yet many wouldn’t even think their software could be part of the solution.

In light of the upcoming Earth Day, we explore how businesses can strive to reduce their carbon footprint through sustainable coding practices.

Why is it important for us to get involved?

Recent research has found that Global data centre electricity use in 2021 accounted for 0.9-1.3% of the total global electricity demand. While many of the large data centre providers such as AWS and Google have introduced ambitious renewable energy or carbon-free targets for powering their data centres, the energy usage of data centres only represents a small portion of the total energy consumed by the running of code and applications. This highlights just how great an opportunity pursuing more efficient coding practices presents when it comes to reducing energy consumption and emissions.

Over the years, our ability to write longer and less refined code with an increasing number of shared frameworks and libraries has greatly expanded our programming capabilities and has been hugely beneficial for driving digital transformation at speed. While it has enabled the modernisation of organisations and industries, we have now reached a critical time where we must start to also consider the efficiency and energy consumption requirements of the code and software we are writing.

This is equally as important for data engineers and scientists to consider. Although they may not write as much software; it is often the implementation of their work that is the most resource intensive.

Key principles of sustainable coding

The principles of green, or sustainable coding should not be considered at odds with existing practices. Rather, they should be incorporated into the principles of engineers to consider when writing and designing code to achieve a balance between functionality and energy usage.

While the pursuit of greener coding practices will be highly tailored to each organisation and each codebase, there are two general considerations that software and data engineers may find useful to keep in mind:

Structural considerations: Encompassing the efficiency and energy measures related to the code blocks.

Behavioural considerations: This is the energy consumption related to the user scenarios, including checking the Twitter feed, sending an email, etc.

By taking time to consider the efficiency and environmental impact of your code, you can start to identify areas in which you will be able to make improvements. For example, an e-commerce business may consider reducing the quality of images uploaded for certain products. This could allow for lower-quality images to be used for less expensive or frequently bought products, to reduce the storage and loading requirements of such images. Thus, providing a balance between user experience and energy consumption.

One of the first and most important steps is simply to become conscious of the environmental impact of certain coding practices, as this will enable you to start including such factors within the design process.

Six tips for sustainable coding

There are many ways to achieve better sustainability through coding practices. We have gathered six tips to help you get started.

1. Keep your code short

Coding will instantly become greener if we adopt a similar development process as we did 20+ years ago when coding was strict to sizes and lengths. Shorter code tends to be less complex and easier to understand, which can result in more efficient execution and easier maintenance over time. While it may take longer to write more concise code, as you need to really understand the problem and find ways to reduce complexity, taking the time to do so is worth it in the long run.

One way to keep your code short is by applying the “Don’t Repeat Yourself” (DRY) principle. DRY coding promotes code reuse by creating modular and reusable functions or modules. By avoiding code duplication and repetition, you can write shorter, more efficient code that is also easier to maintain and update.

2. Introduce Code Profiling and Query Analysers

Introducing code profilers and query analyzers is a great way to help make your code more efficient and sustainable. These tools can help you identify performance bottlenecks, improve resource utilization, and reduce the risk of bugs and errors.

Code profilers are tools that help you measure the performance of your code by tracking how long each function or line of code takes to execute. By using a profiler, you can identify which parts of your code are taking the most time and resources and optimize them accordingly. For example, you may discover that a certain loop or function is taking much longer to execute than it needs to, and you can then optimize the code to reduce its processing time.

Query analyzers are tools that help you analyze database queries and identify ways to improve their efficiency. By examining query execution plans and resource usage, query analyzers can help you identify performance issues and optimize your database queries for better performance.

3. Avoid software bloat by minimising open source and imports

A large proportion of software developments leverage open-source code. While the use of general libraries with wide varieties of functionality can be useful for first developing applications, this can lead to software bloat. While it is convenient to use pre-existing code libraries and packages, it is important to be selective about what you use and to only import what is necessary. Every imported library adds to the size of the application, which can increase its loading time and decrease its performance.

By minimizing the use of open-source libraries and third-party imports, you can reduce the amount of code in your project and improve its performance. Additionally, importing code from unknown sources or outdated packages can pose security risks. In fact, we discussed the risks of using open-source software and how to mitigate them in our recent blog.

Code scanning tools can be used to analyse the size and complexity of code bases, helping identify areas of code that are bloated and could be optimised. By minimizing the use of open-source and imports and using code scanning tools, you can create leaner and more sustainable code that is also easier to maintain and scale over time.

4. Tackle technical debt

Technical debt, also known as tech debt or code debt, describes what results when development teams take actions to expedite the delivery of a piece of functionality or a project which later needs to be refactored. In other words, it’s the result of prioritising speedy delivery over perfect code.

Such code represents an opportunity for efficiency improvements, as it is likely optimisations can be made through refactoring the code. Tackling technical debt can help to both improve the product and reduce its carbon footprint.

5. Migrate to a green hosting provider

Migrating your applications and computing resource to the cloud, in particular a green hosting provider, can be a quick win in terms of reducing the carbon footprint of your IT estate. Cloud computing resources are often newer and more efficient, with large data centres also providing energy efficiency gains due to economies of scale. Furthermore, while maintaining compliance with data protection guidance, relocating your servers to colder climates where possible can reduce Greenhouse Gas (GHG) emissions by up to 8%.

6. Consider using a more energy-efficient programming language

While the use of easier-to-read languages such as Python can be extremely useful for producing proof of concepts for products and initial research for models, different programming languages vary significantly in their efficiency. A team of researchers from Portugal studied the 27 most used programming languages to determine their efficiency and energy consumption.

From this research, they found the 5 most efficient programming languages were:

• C

• Rust

• C++

• Ada

• Java

They also found that faster programming languages are not always the most energy efficient. While Java ranks well in both energy efficiency and speed, Python, Perl, Ruby, etc., are among the least efficient programming languages.

Conclusion

As the climate crisis deepens and the pressure on businesses to become more sustainable grows, technology and coding practices shouldn’t be ignored.

Adopting greener coding practices can be an excellent way of helping your business play its role in reducing carbon emissions, helping to contribute toward sustainable IT, and conserving the environment.

More from our Emerging Tech Series:

• What is low code or no code development? (Emerging Tech Series: Part 5)
• Explained: Bug Bounty Testing (Emerging Tech Series: Part 4)
• Explained: Minimum Viable Architecture (Emerging Tech Series: Part 3)
• Explained: Machine Learning Code Generation (Emerging Tech Series: Part 2)
• Cloud Sustainability (Emerging Tech Series: Part 1)

Thanks to Data Scientist Jack Foster and Technical Architect Mark Hastry for their contributions to this piece.