How to prioritise sustainable operations

David Gammie, CTO at iomart, outlines some methods to improve the sustainability and efficiency of today’s data centres.

Data centres are incredibly energy intensive systems. In fact, they are estimated to be responsible for up to 3% of global electricity consumption today, and are projected to increase to 4% by 2030. We live in a world where everything is now stored digitally and with the UK’s net zero goals, it’s vital that organisations seek out innovative ways to reduce energy consumption of that data processing/storage and work to be more sustainable. 

Every business uses and stores data to perform business processes and provide management information. As such, data centres are a crucial part of daily operations within an organisation. Unfortunately, however, data centre sustainability remains a challenge, particularly when it comes to the cloud industry. Operators must balance sustainability goals and meeting net-zero targets with the continuous, fast-paced digitisation – which is no easy feat. 

While the adoption of AI and IoT devices has increased the demand on data centre infrastructure, which needs additional power, the new technologies can also help IT leaders streamline their operations. Ultimately, understanding how to utilise these new solutions to boost productivity within data centres will translate to lower emissions, as operators will be able to filter out redundant applications and processes and improve overall efficiency. 

Making data centres more sustainable

Organisations need to make sustainability a more embedded and higher profile criteria for key decisions on how data centres are built and populated; they need to be sustainable by design. Increasingly, organisations also need to plan for the rapidly changing weather patterns. 

In order to ensure efficient and sustainable operations within their data centres, IT and I&O leaders should use real-time and predictive analytics in order to maximise the life cycle of their data centre assets and not prematurely replace infrastructure, such as cooling systems, backup generators or power supplies. 

One big change data centre operators can make is switching to renewable energy. Wind and solar power is already growing in popularity around the world, particularly for data centres. In fact, many operators are combining energy storage systems with their renewable energy systems in order to ensure a flow of constant power and ensuring servers maintain their uptime.

Organisations could also choose to invest in GHG emission offsetting programmes, which include recycling materials to extend their lifespan, refurbishing and recovering valuable and environmentally sensitive materials, and implementing a circular economy programme to scale back e-waste. 

Water also plays an especially big part in data centres, as it’s used for cooling. As such, organisations should consider using water usage effectiveness (WUE) in order to track all sources of water, as well as using grey water sources, such as seawater, recycled water or rainwater collection systems, instead of potable mains. 

Another step that can be taken is looking to take advantage of colocation options. This can help organisations reduce or close any internal data centres that aren’t being utilised to their full extent, or those that are located in areas where renewable or ‘greener’ energy isn’t as accessible. 

Additionally, Gartner recommends that leaders establish their IT sustainability goals and measure their progress in order to track the reduction of their emissions and the progress they’re making. This way they can better understand whether their approach is working, or they need to change up their methods. 

Optimising at the migration stage

While cloud providers themselves are responsible for their own emissions, there’s a core change that needs to happen with how solutions are deployed, designed, and retired within organisations if not in use. This can include optimising software development and the processing to make it as efficient as possible and appropriately managing infrastructure so it runs optimally. Taking these measures can significantly reduce the amount of energy needed to power solutions and demonstrates how a little goes a long way. 

When it comes to transitioning to the cloud, businesses should be assessing which applications are needed and which can be retired, as this also decreases the power needed alongside the carbon footprint. For instance, any applications that aren’t being used regularly within the company, as well as full email inboxes, should be sorted through in order to reduce the load that is being transferred. 

With this, IT leaders should also modernise and optimise applications during the migration stage. Not a ‘lift and shift’ but a ‘transform and move’ mindset. This can minimise storage requirements and also helps businesses do more with their data, such as integrating AI and machine learning for automation and wider benefits. 

Integrating AI into data centre operations

In order to remain competitive and streamline processes, organisations can integrate AI and machine learning (ML) into their technologies. In doing so, they can create an autonomous data centre that optimises and performs data engineering tasks without human intervention. 

Outages are also becoming more frequent and costing data centres significant downtime and funds, in addition to wider business and brand impact. Using AI to automate certain processes, IT leaders can noticeably minimise downtime and improve their system’s reliability and efficiency. 

From a sustainability perspective, AI can help reduce the amount of hardware IT leaders purchase, manage and monitor, which will consequently lessen energy consumption. An AI solution can help businesses to better scale their resources to avoid congestion, and help managers to oversee their resourcing levels, make real-time decisions and improve overall efficiency. 

That being said, operators need to be aware that solutions such as AI may require more power to run. Therefore, before blindly implementing an AI or ML model, it’s imperative that businesses identify and understand the right use cases for these solutions, train their workforce on how to use it responsibly, and invest in the necessary infrastructure to run it efficiently. 

Making a difference

Ultimately, the future of data centres should involve working towards feasible sustainability goals and putting effort and investments into greener operations. 

Cloud providers in particular should be creating that sustainability vision, driving a roadmap and making adjustments, no matter how small, in order to progress on their net zero journey and reduce their carbon footprints wherever possible. 

Related Articles

Top Stories