A new report from Seagate Technology Holdings plc, a mass-capacity data storage provider, highlights the growing sustainability challenges facing data centres as organisations scale to meet the demands of AI.

Goldman Sachs Research predicts a substantial increase in global power demand from data centres, potentially rising by as much as 165 % by 2030, compared with 2023 levels. Seagate ' s report corroborates this, indicating that energy usage is now a top concern for 53.5 % of business leaders. The confluence of rapidly increasing data volumes, diminishing power efficiency gains, and the widespread adoption of AI is compelling organisations to address carbon emissions, infrastructure expansion and total cost of ownership( TCO) simultaneously.

The report also highlights a disconnect in life cycle management. While 92.2 % of respondents acknowledge the importance of extending the life cycle of storage equipment, only 15.5 % consider this a primary factor in their purchasing decisions.

“ Data centres are under intense scrutiny – not only because they support modern AI workloads, but because they are becoming one of the most energy-intensive sectors of the digital economy,” said Jason Feist, Senior Vice President of Cloud Marketing, Seagate.

“ This calls for a fundamental shift in how we think about data infrastructure – not as a trade-off between cost and sustainability, but as an opportunity to optimise for both.”

The report outlines the challenges organisations face as they seek to expand their data capabilities. They are presented with three primary options: improving efficiency within their existing infrastructure, expanding their data centre footprint, or migrating workloads to the cloud.

Three strategies for a sustainable data centre industry

The Decarbonizing Data report proposes three strategic pillars for building a more sustainable data future. Firstly, advances in computational power, storage areal density and energyefficient technologies, such as liquid and immersion cooling and improved HVAC systems, can play a crucial role in significantly reducing energy consumption and carbon emissions.

Growing demand for data storage due to AI

A significant majority, 94.5 % of respondents, reported increasing data storage needs, with 97 % anticipating that the growth of AI will further exacerbate this demand. While there is a high level of concern regarding environmental impact, with nearly 95 % of respondents expressing concern, this concern does not translate into purchasing decisions.

Only 3.3 % of respondents prioritise environmental impact when making purchasing choices. The report identifies several key barriers hindering the adoption of more sustainable practices within data centres. These include high energy consumption( 53.5 %), the raw material requirements of data centre infrastructure( 49.5 %), physical space constraints( 45.5 %), infrastructure costs( 28.5 %) and acquisition costs( 27 %).

Secondly, practices such as refurbishing, reusing and maintaining storage equipment to extend its lifespan and reduce waste are crucial, as well as real-time environmental monitoring and transparent reporting to encourage accountability across the data centre environment.

Thirdly, the report notes that achieving meaningful emissions reduction, encompassing Scopes 1, 2, and 3, as defined by the Greenhouse Gas Protocol, requires collaboration across the entire value chain. This includes collaboration between vendors, suppliers and cloud service providers.

" Sustainability cannot be solved in isolation. A holistic approach spanning infrastructure, life cycle management and industrywide accountability could ensure that the growth of AI and data centre operations does not come at the expense of the environment," added Feist. �

INTELLIGENT POWER AND COOLING www. intelligentdatacentres. com 59