Data centres have become a central character in Asia’s digital transformation story in 2025. The region’s data centres have been experiencing a surge in demand for capacity and computing power, catalysed by enterprises' adoption of artificial intelligence (AI) technologies. According to the Asia Pacific Data Center Market Report, the data centre industry is expected to be worth US$71.67 billion by 2032.
Data centres are critical for businesses to address new infrastructure challenges that are emerging as AI continues to evolve in 2025. New large language models are entering the market, democratising access to AI and potentially creating even stronger demand. Asia’s regulatory landscape is likely to evolve as nations prioritize data sovereignty and security. Sustainability targets mean that there is a need to maximise the energy efficiency of the infrastructure that powers AI. Businesses must also rethink the location and anatomy of their IT infrastructure to meet requirements that will continue to transform in the coming years.
Asia – A global data centre Hub
The global scales are tilting towards Asia as a primary data centre hub. The combined power capacity of the region’s data centres is forecasted to reach 94.4 gigawatts (GW) in the next three years.
In particular, Southeast Asia’s digital ecosystem is projected to grow to nearly US$1 trillion by 2030. With over 400 million internet users and a 70% internet penetration rate in most of its countries, digital sectors like e-commerce, online media and ride-hailing services are thriving. The sub-region has also seen accelerated growth with the adoption of cloud technologies and national digitization projects, such as Singapore’s refreshed Smart Nation 2.0 and Malaysia’s MYDigital policies.
Singapore boasts over 1.4 GW of data centre capacity, and has established itself as a regional powerhouse. The lower costs of energy and land in emerging markets, such as Malaysia, Indonesia and Vietnam, are also attracting data centre investment. This growth will also serve the rising computational and data storage demands of new digital services and solutions in the region.
See also: Embracing the cloud-native and AI transformation journey
Infrastructure, transformed
Enterprise data requirements are evolving in response to rapid technological advancements and a broadening AI landscape. As businesses increasingly rely on multiple AI models working together, they will require AI-ready data centre campuses that offer proximity and interconnectivity to data sources.
The growing adoption of AI inference-enabled solutions will also change the way businesses design and manage digital infrastructure, making it more data-centric. This will reduce an enterprise’s centralieed data load, leading to a focus on data aggregation and analysis, as data processing locations are placed closer to customers and end-users.
See also: Is your cloud more fortress or folly? Reinforcing cloud security in the digital age
Hybrid infrastructures – combining public and private data – are also on the rise in this region. Data centres act as hubs for low-latency interconnection, bringing clouds and users closer to data, wherever it resides.
A focus on data sovereignty
Driven by security and regulatory concerns, data sovereignty has become a focus for many of Asia's enterprises. According to the Global Data Insights Survey: The State of Data and AI in Asia Pacific, 77% of companies in Asia Pacific have already adopted a distributed data approach.
This means the traditional borderless cloud model, once a cornerstone for multinationals across Asia, is facing a significant challenge. Companies that previously relied on the cloud's instant scalability and seamless integration now must manage a patchwork of sovereign clouds, each tailored to unique local requirements. This shift not only increases operational costs and data management complexity but also introduces new compliance risks that can hinder innovation and expansion.
Asia’s data centre sector is playing a critical role in helping businesses meet these challenges. These data centres act as connected hubs, enabling companies to store and process data within specific geographical boundaries, adhering to data sovereignty laws while still leveraging the power of the cloud.
Accounting for the energy efficiency of AI
For AI to reach its transformative potential, its environmental impact will need to be accounted for, especially relating to energy and water consumption. In 2024, the global AFCOM State of the Data Center Report revealed that a single rack consumes an average of 12 kilowatts (kW) of power, almost double the amount from two years ago. Data centre water consumption has also increased to help facilities cool down the excessive heat produced by AI workloads.
To stay ahead of the latest tech trends, click here for DigitalEdge Section
Thankfully, sustainability remains at the top of the business agenda, with 59% of businesses citing sustainability goals as the most important factor impacting AI strategy. To ensure this, data centre facilities have adopted innovative methods of reducing energy and water consumption. For example, Digital Realty’s award-winning use of direct current ionisation electrolysis in its water cooling systems has saved 1.24 million litres of water monthly and is using 100% renewable energy in its Singapore-based facilities.
Future-ready infrastructure
Enterprises will likely double down on AI adoption in the coming years and rely on the technology’s inference capabilities to complete complex tasks or generate accurate predictions. Inference workloads will process less data, but generate output more quickly to deal with highly specific tasks.
To adapt, enterprises should look for data centre solutions that combine state-of-the-art technology with pervasive global networks to maximise efficiency and performance and enhance access to a diverse AI ecosystem. These data centres should act as core connectivity hubs and provide efficient meeting places for enterprises and service providers to innovate with AI.
The year ahead will promise changes for business IT. Energy efficiency, data sovereignty, regional growth and infrastructure demands will be developments that enterprises need to grapple with in the age of AI. The data centres they choose will be their key ally in this story.
Serene Nah is the managing director and head of Asia Pacific at Digital Realty
