The amount of data in existence today is inevitably placing huge amounts of pressure on data centre facilities. Mark Molyneux, EMEA CTO at Cohesity, discusses ESG research, commissioned by Cohesity, which explores data growth rates at more than 3,000 organisations reported that: 39% of the companies surveyed saw data volumes increase by up to 20% each year, 31% said they grew up to 50% annually and 28% saw their data swell by more than 50% each year. Data consumers are contributing to these numbers and are happy to let it grow, so long as they can access it for their needs. “And herein lies the challenge faced by data centre leaders; how do you control that data growth, without impacting consumer requirements?” questions Molyneux.
He says the answer lies in compliance and regulation, underpinned by security. “Data, often in fragmented silos spread across their environments, has limited rules in place for retention and this is where we see the growing mountain of uncontrolled data causing pressure on data centre leaders. They are continually forced to scale platforms up/out and even forced into moving data off premise in an attempt to ease capacity at their sites.”
Molyneux believes that introducing a solid set of compliance rules will greatly aid data centre leaders in controlling growth and financial spend and aid their presentation of data to consumers. “Every company will have a Relevant Record Strategy that sets out what it needs to retain and for how long,” Molyneux continued. “Companies that have these in place are very good at applying the rules to physical paper or to applications, but typically fall short when it comes to stored data. Many firms do not classify their data in line with the record strategy and this is how the data mountain grows uncontrollably. When data is classified, it makes it very simple for technology to solely retain relevant data for a defined period. By introducing rules of control you start to positively impact your data footprint and become more sustainable in the use of both technology and locations, and of course you are fully compliant to regulations.”
I wanted to glean further insight into how the pressure on data centres can be relieved, so I asked the experts…
Dean Boyle, CEO, EkkoSense
The simple answer is I’m not sure there’s any way now to persuade people to use less data. As consumers we’ve all grown addicted to using the many different online services that are the key driver of global data growth – and our collective data habit shows no signs of slowing down.
Statista research shows that the volume of data generated worldwide each year has quadrupled since 2018, with 2020 – the year of global lockdowns – seeing 56% annual growth alone. Video now accounts for over half of global data traffic and it’s easy to see how. We used to watch YouTube videos at a 480p resolution – that works out at around 500 MB an hour. Now though, most of us watch at 1080p or 3 GB an hour, while 4K actually eats up as much as 16 GB for an hour’s hi-res viewing. And with billions of people using YouTube, or services such as Netflix, you quickly begin to add up to the 120 zettabytes of data that are expected to be generated this year.
So if consumers aren’t prepared to ease up on video streaming or stop using social media services such as Facebook, Instagram, Snapchat and TikTok, and if organisations won’t slow down on their Digital Transformation initiatives, how can we relieve the pressure on data centres? And perhaps more importantly, is there anything data centres can do to reduce their energy usage?
At EkkoSense we believe that addressing data centre cooling efficiency needs to be a key area of focus for operations teams. It’s one of the most effective ways to optimise performance and secure environmental improvements – particularly given that cooling data centres can account for up to 40% of their total energy consumption. We estimate that the global data centre industry is currently overspending on cooling energy costs by over £1.5 billion each year.
That’s why the EkkoSense team focuses on making it as easy as possible for operations teams to gather and visualise their cooling, power and space data at a much more granular level. Using EkkoSoft Critical, our 3D visualisation and analytics platform, teams can visualise airflow management improvements, manage complex capacity decisions and quickly highlight any worrying trends in cooling performance.
Having access to this kind of insight can make a huge difference. Our research found that data centre cooling utilisation currently averages just 40% across the industry. Effective data centre optimisation helps release this expensive cooling capacity, helping organisations to handle increased workloads and protect against unnecessary spending on additional cooling. Using this approach, we’ve also been able to reduce data centre cooling costs by an average of 30%, in turn helping to unlock immediate carbon savings that can contribute directly to corporate net zero programmes.
Julian Boneham, Data Practice Director at Node4
Data is growing exponentially as businesses are generating data at a higher pace and from more sources than ever. In fact, the volume of data created, captured, copied and consumed worldwide is predicted to grow 51% by 2025 with 181 zettabytes of data expected to be generated globally. This creates a challenge for businesses who become overwhelmed with data and can no longer determine exactly what is relevant in terms of making informed business decisions. Many organisations have, therefore, previously decided to store it all, rather than risk deleting data that may be useful in the future, inundating data centres causing huge pressure on their storage capacity.
Yet, businesses are becoming increasingly aware of the environmental impact of large data footprints, resulting in a change of attitude. Many are no longer willing to store all of their data and are taking the opportunity to better analyse what data needs to be stored and what can be removed. Not only is this more sustainable and helps reduce an organisation’s carbon footprint, but crucially it frees up space in data centres and is a key factor in encouraging consumers to reduce the pressure on these organisations.
While these are external benefits, it is also advantageous for the organisation. Tying it in with a wider data strategy, businesses can encompass data architecture and analytics requirements to determine the optimal volume of data that needs to be stored and from which data sources. This will enable organisations to discover what data they have, where it is stored, why it is valuable (or not!) and if it needs to be stored or can be discarded.
It is also a great opportunity to clean-up legacy data that businesses probably didn’t even realise they were still storing. This could be for outdated or obsolete systems, formats, or technologies that are often difficult to access and are no longer needed or used by the organisation. I’m sure many businesses would be surprised to see how much data they are storing unnecessarily.
Not only will this free up space in overloaded data centres but it can also reduce costs. After all, every business will pay for space within the data centre so making sure that you are only storing the information that is business-critical will significantly reduce costs. In the current economic climate where every penny counts, this is a key element to persuade consumers to reduce pressure on data centres as clearing out any unused or unwanted data can make a huge difference financially.
Matt Watts, Chief Evangelist, NetApp
No-one is debating the value of data. The way we generate and consume data isn’t going to disappear, so for businesses, the challenge is to find ways in which to eliminate waste and save costs where possible. But consumers must do their part too.
One simple way to this is by addressing our relationship with devices. Every email sent creates around 0.5g of carbon dioxide equivalent (CO2e). If you add a couple of attachments, that number can quickly increase to 10–20g CO2e. It’s the same for Facebook, Instagram, TikTok and for pretty much anything you interact with through your device of choice.
Every action in the digital world has an impact on the environment, the consequences of which are often out of sight and out of mind. Many will have seen footnotes in emails with reminders of the environmental consequences of printing them or their attachments but given the impact on the environment of sending an email, maybe it’s time to consider a digital detox.
Instead of sending an email with a big attachment to 200 people, just send it to the few people who actually need to see it. You’ll be doing everyone a favour and you’ll save a little bit of CO2e, too.
So what about businesses? If they create and follow a data strategy built around data minimisation, they are only storing the data they need, while also being one step closer to meeting Environmental, Social Governance (ESG) targets.
According to the Data Waste Index report, which was released recently by NetApp, a worryingly high 41% of data currently stored by UK organisations is unused or unwanted. This data will never be used again, often because companies haven’t worked out how to gain value from it, or simply because it’s just data that serves no further purpose beyond that for which it was created.
Clearly then, businesses need to know how much data they have, what the data is and where it is located. Visibility across your data landscape is essential whether on-premises, or in hybrid cloud and cloud environments. This requires effective data management software which can work across all data siloes in order to understand how much data they can reduce. From this action, businesses have the ability to remove data that is clearly unnecessary, inactive or duplicated.
In many cases storing data in the cloud for production, development, archive, or backup can improve the situation, and drawing on the expertise of vendors and partners who can offer the tools, analytics and understanding will help you to make better environmental decisions. With help, organisations and consumers alike can reduce both their energy cost exposures and their carbon emissions.
