E01 What Is a Data Centre

Key Takeaways

• A data centre is a critical infrastructure facility that provides the computing services behind cloud computing, artificial intelligence and most modern digital services.
• Although digital services appear virtual, they depend entirely on physical infrastructure that is increasingly being built in Malaysia to serve users across the region and the world.
• Modern data centres have evolved from small corporate server rooms into large-scale infrastructure requiring sophisticated electrical, cooling, networking, security and control systems.
• The industry’s pursuit of near-continuous reliability shapes almost every aspect of data centre design, operation and investment.
• The same engineering systems that enable reliable digital services also give rise to challenges such as energy demand, water use, heat rejection, noise and land requirements.
• As Malaysia becomes a regional data centre hub, sustainable success depends on balancing technological progress, economic development, environmental stewardship and community wellbeing.
• Understanding data centres requires a systems perspective that considers their engineering, regulatory, economic, environmental and social dimensions together.

The Invisible Infrastructure Behind Everyday Digital Life

Every day, billions of people send messages, watch videos, make digital payments, store photographs, attend online meetings and increasingly interact with artificial intelligence. These activities feel almost effortless. A video begins playing within seconds. A bank transfer completes instantly. An AI assistant responds in natural language. From the perspective of the user, these services appear almost magical—existing somewhere in an invisible “cloud.”

In reality, there is nothing virtual about the physical infrastructure that makes these services possible.

Every digital request begins a journey through fibre-optic cables, communication networks and computing systems located in buildings somewhere in the world. A message sent from Kuala Lumpur may be processed by computing equipment in Johor. An online purchase made in Singapore may be authenticated by servers in Cyberjaya. An artificial intelligence model assisting a researcher in Europe may perform part of its computation inside a data centre located in Malaysia.

As our economies become increasingly digital, the infrastructure supporting them is becoming increasingly physical.

This is one reason why data centres have become one of the fastest-growing forms of critical infrastructure worldwide. While they often appear externally as simple industrial buildings, they are in fact highly sophisticated computing facilities that enable governments, businesses and individuals to communicate, transact, innovate and increasingly create new forms of knowledge through artificial intelligence.

Malaysia is becoming an increasingly important participant in this global transformation. New data centres being developed in Johor, Cyberjaya and the Klang Valley are not intended solely to serve Malaysian users. Many form part of regional and global cloud platforms supporting businesses and consumers throughout Southeast Asia and beyond. In other words, infrastructure built within Malaysian communities increasingly provides digital services to the world.

MDCO Insight: Although digital services appear borderless, every digital interaction ultimately depends upon physical infrastructure located within real communities somewhere in the world.

From Office Computer Rooms to Global Digital Infrastructure

The idea of housing computers in dedicated facilities is not new. Large organisations have maintained computer rooms since the early days of commercial computing in the 1960s and 1970s. Banks, universities, government agencies and multinational corporations often allocated secure areas within their office buildings to house expensive mainframe computers responsible for payroll, accounting and record management.

These early facilities resembled specialised office infrastructure rather than industrial developments. They existed to support the internal operations of a single organisation, much like meeting rooms, archives or administrative offices. Their scale was relatively modest because the number of users was limited and computing power remained expensive.

The emergence of the internet fundamentally changed this relationship.

Instead of computers serving only employees inside a single organisation, they increasingly began serving customers, suppliers and eventually billions of users connected through the internet. Email moved online. Websites became interactive. Businesses adopted cloud software rather than installing programs on individual computers. Streaming replaced physical media. Smartphones became powerful digital interfaces connected continuously to online services.

Perhaps the most profound transformation has occurred in recent years with artificial intelligence. Rather than relying primarily on the processing capability inside personal computers or smartphones, increasingly sophisticated AI models operate on enormous clusters of specialised processors housed inside large data centres. The device in a user’s hand has become less a standalone computer and more a gateway to computing resources distributed across global networks.

As demand expanded from hundreds of users to millions—and eventually billions—the supporting infrastructure also changed dramatically.

Server rooms evolved into purpose-built facilities.

Individual buildings evolved into campuses.

Enterprise computer rooms evolved into hyperscale data centres capable of supporting cloud platforms serving entire regions.

The role of the data centre also changed. Rather than supporting a single company, it became part of society’s shared digital infrastructure, much like airports, ports, highways or electricity networks. Although the services they provide are digital rather than physical, modern economies increasingly depend upon them in much the same way.

This evolution explains why discussions surrounding data centres have also changed. They are no longer viewed solely as information technology assets but increasingly as strategic national infrastructure with implications for economic development, energy planning, environmental management and international competitiveness.

MDCO Insight: Data centres have evolved from supporting individual organisations to supporting the digital infrastructure upon which modern economies increasingly depend.

How a Data Centre Works

Many people assume that when they use a smartphone or personal computer, most of the computing takes place inside the device itself.

That assumption was largely true several decades ago. Today, however, it is only partly correct.

When someone searches for information online, uploads photographs to cloud storage, watches a streaming service, books a flight, completes an online banking transaction or interacts with an AI assistant, much of the actual computation occurs somewhere else. The user’s device primarily captures the request, displays the results and provides an interface through which people interact with digital services.

The intensive processing often takes place remotely inside data centres.

Consider a simple example.

A user asks an AI assistant to summarise a document. Within fractions of a second, the request travels through telecommunications networks to a remote data centre. There, thousands of specialised processors work together to interpret the request, retrieve relevant information, generate a response and send the results back across the network. The entire process may involve multiple computing systems operating in different geographic locations before the answer appears on the user’s screen.

This model is known as cloud computing, although the term “cloud” can sometimes be misleading. There is no cloud in the sky performing these calculations. Instead, the “cloud” refers to a globally distributed network of physical data centres connected by high-speed communication infrastructure.

This approach offers several important advantages.

Rather than requiring every smartphone or laptop to contain enormous computing power, resources can be shared among millions of users. Software can be updated centrally. Data can be protected through multiple backup locations. Computing capacity can expand rapidly as demand increases. Organisations can access world-class computing resources without constructing their own facilities.

Artificial intelligence has accelerated this trend even further.

Training and operating advanced AI models requires computing resources far beyond the capability of ordinary personal devices. Modern AI systems rely on thousands—or sometimes tens of thousands—of graphics processing units (GPUs) working together simultaneously. Housing, powering and cooling this equipment economically is only practical within specialised data centres designed specifically for this purpose.

As a result, the world’s digital future increasingly depends not only upon better software but also upon better physical infrastructure capable of supporting ever larger computing workloads.

MDCO Insight: Increasingly, our phones and computers function less as standalone computing devices and more as intelligent gateways connecting people to vast shared computing infrastructure.

Scaling Up a Computer into Critical Infrastructure

One useful way to understand a modern data centre is to imagine taking the components inside a personal computer and scaling them up thousands—or even hundreds of thousands—of times.

Every computer contains processors that perform calculations, memory that stores temporary information, storage devices that retain data, power supplies that convert electricity, cooling systems that remove heat and communication components that exchange information with other devices.

A data centre contains exactly the same functional elements.

Instead of one processor, it may contain hundreds of thousands.

Instead of one cooling fan, it operates extensive mechanical cooling systems.

Instead of plugging into a household electrical socket, it may connect directly to high-voltage transmission networks operating at 132 kV or 275 kV.

Instead of one internet connection, it connects simultaneously to multiple national and international fibre-optic networks.

However, increasing scale introduces engineering challenges that do not exist inside ordinary computers.

Every calculation performed by a processor generates heat. A single laptop produces only a small amount, easily removed by a fan. Multiply that by tens of thousands of servers operating continuously, and the resulting heat becomes an industrial engineering problem requiring sophisticated air-conditioning systems, chilled water plants, cooling towers or increasingly liquid-cooling technologies.

Similarly, a household computer can simply be restarted after a power interruption. Modern cloud services cannot.

Many digital services now support hospitals, emergency communications, financial systems, airports, manufacturing facilities and governments. Interruptions measured in seconds can affect millions of users simultaneously. Consequently, data centres incorporate multiple independent power supplies, battery energy storage systems, backup generators and sophisticated monitoring systems designed to detect abnormalities before failures occur.

Networking infrastructure also grows dramatically in complexity. Thousands of servers must exchange enormous volumes of information continuously while maintaining extremely low communication delays. Extensive fibre-optic networks, switching equipment and intelligent control systems therefore become integral parts of the facility rather than optional additions.

Supporting all these systems requires an equally sophisticated layer of automation. Modern data centres continuously monitor electrical loads, cooling performance, equipment health, fire protection, physical security, environmental conditions and network traffic through integrated building management and infrastructure monitoring systems.

At this scale, a data centre is no longer simply a building filled with computers.

It becomes a highly integrated infrastructure system where electrical engineering, mechanical engineering, information technology, telecommunications, automation, architecture and environmental management operate together to deliver one outcome: reliable digital services.

The more society depends upon those services, the more important every supporting system becomes—not only for performance and reliability, but also for efficiency, sustainability and responsible coexistence with surrounding communities.

MDCO Insight: A hyperscale data centre is not merely a large computer—it is a complex infrastructure ecosystem in which every engineering system exists to support reliable digital services while balancing performance, sustainability and community impact.

Why Reliability Shapes Everything

If there is one principle that distinguishes a data centre from almost every other type of building, it is the expectation that it should continue operating regardless of what happens around it.

Most buildings can tolerate occasional interruptions. A shopping mall may close temporarily during a power outage. An office building can suspend operations until electricity is restored. Even manufacturing facilities often have planned shutdowns for maintenance.

Modern digital services, however, increasingly operate under different expectations.

People expect online banking to remain available throughout the day. Airlines rely on continuous reservation systems. Hospitals depend upon uninterrupted access to patient records. Emergency services, logistics networks, cloud platforms and increasingly artificial intelligence services all assume that the underlying digital infrastructure will remain available every second of every day.

This expectation has fundamentally shaped the engineering of modern data centres.

Instead of relying upon a single electricity supply, data centres typically incorporate multiple independent sources of power. Incoming electricity from the transmission grid is distributed through redundant electrical systems. Uninterruptible Power Supply (UPS) systems bridge even the briefest interruption while standby generators start automatically should utility power fail. Critical equipment is often duplicated so that maintenance or component failure does not interrupt operations.

Cooling systems are designed with the same philosophy. If one chiller, cooling tower or pump fails, another immediately takes over. Network connections are similarly duplicated through multiple fibre routes and communication providers to reduce the likelihood that a single incident can disconnect the facility from the outside world.

The industry has developed internationally recognised standards to describe different levels of resilience. The widely referenced Uptime Institute Tier classifications provide a common framework for comparing the fault tolerance and maintainability of different facilities. At the same time, operators commonly express performance in terms of availability, with leading facilities targeting “five nines” availability, or 99.999% uptime, equivalent to only a few minutes of unplanned downtime each year.

Achieving this level of reliability is neither simple nor inexpensive. Every additional layer of resilience requires additional equipment, more floor space, greater electrical capacity and increasingly sophisticated monitoring systems. Reliability therefore becomes not only an engineering objective but also a major driver of investment, operational complexity and resource consumption.

This relationship helps explain why discussions surrounding data centres often involve apparent contradictions. The same backup generators that improve resilience may also become sources of noise during routine testing. Additional cooling equipment that protects computing systems may increase energy consumption or require greater water use. Security measures designed to protect critical digital infrastructure may also make facilities appear closed or inaccessible to surrounding communities.

Seen from this perspective, many of the features that attract public attention are not independent design choices. They are consequences of a single overriding objective: maintaining continuous digital services upon which modern society increasingly depends.

MDCO Insight: Reliability is not simply one design requirement among many; it is the principle that shapes almost every engineering decision within a modern data centre.

Balancing Performance, Sustainability and Community

As data centres have grown larger and more powerful, expectations have expanded beyond reliability alone. Society increasingly expects digital infrastructure to deliver high performance while also operating efficiently, minimising environmental impacts and coexisting responsibly with surrounding communities.

Meeting these expectations requires balancing objectives that are sometimes complementary and sometimes in tension.

For operators and customers, the primary objective is to provide secure, high-performance computing services capable of responding instantly to millions of users. Investors seek facilities that remain commercially competitive over decades. Governments view data centres as strategic infrastructure supporting national digital transformation and economic growth. Local communities naturally focus on how these developments affect their neighbourhoods, while regulators seek to ensure that public health, safety and environmental standards are maintained.

These perspectives are all legitimate, yet they do not always point in exactly the same direction.

Engineers therefore devote considerable effort to improving efficiency without compromising reliability. One commonly used indicator is Power Usage Effectiveness (PUE), which compares the total electricity consumed by a facility with the electricity used directly by computing equipment. A lower PUE indicates that a greater proportion of energy supports productive computing rather than auxiliary systems such as cooling or lighting.

Water Usage Effectiveness (WUE) provides a similar measure for facilities using water-based cooling technologies. Advances in liquid cooling, more efficient chillers, artificial intelligence-assisted workload management and renewable energy integration all aim to reduce the environmental footprint of increasingly powerful computing infrastructure.

Yet improving efficiency alone does not necessarily reduce total resource consumption.

Global demand for digital services continues to grow rapidly. Artificial intelligence, high-definition video, cloud computing and connected devices require ever greater computing capacity. Consequently, even as individual facilities become more efficient, the overall demand for electricity, land, water and supporting infrastructure may continue to increase.

Many of the environmental and community impacts associated with data centres arise directly from the engineering systems discussed earlier. Cooling systems reject heat generated by computing equipment. Backup generators produce emissions during testing or emergency operation. Electrical infrastructure requires high-voltage substations and transmission connections. Large campuses occupy substantial areas of land and may influence local transport, drainage and utility planning.

Recognising these relationships is important because it shifts discussion away from viewing environmental impacts as isolated problems. Instead, they become part of a broader systems challenge: how can society continue expanding digital infrastructure while reducing its demands on natural resources and neighbouring communities?

Increasingly, innovation is directed precisely at answering this question.

MDCO Insight: The future of data centre development depends not on choosing between digital progress and environmental stewardship, but on improving the way both are achieved together.

Malaysia’s Place in the Global Digital Economy

Malaysia has participated in the development of digital infrastructure for many years, with Cyberjaya emerging as one of the country’s earliest purpose-built technology hubs. Enterprise data centres supporting banks, telecommunications providers and government agencies have long formed part of the national digital landscape.

The scale of development witnessed today, however, represents something fundamentally different.

Driven by rapid growth in cloud computing, artificial intelligence and regional digital demand, Malaysia is emerging as one of Southeast Asia’s principal destinations for hyperscale data centre investment. Johor has attracted large international developments supported by its strategic location, land availability and strong connectivity to regional markets. The Klang Valley continues to expand as an important enterprise and cloud services hub, while new investments are extending digital infrastructure into other parts of the country.

Much of this growth is driven by regional rather than purely domestic demand.

Businesses throughout Asia increasingly rely on cloud platforms operated by global technology companies. As neighbouring markets encounter constraints related to land availability, electricity capacity or planning limitations, Malaysia has become an attractive location from which these services can continue expanding across the region.

This changing role carries both opportunities and responsibilities.

Economically, data centres contribute investment, employment, technical expertise and supporting industries ranging from engineering consultancy to construction, utilities and facility operations. Strategically, they strengthen Malaysia’s position within the regional digital economy and may support broader ambitions in artificial intelligence, advanced manufacturing and digital services.

At the same time, these facilities become part of the physical landscape of Malaysian communities. Their success therefore depends not only upon technical excellence or commercial viability but also upon responsible planning, transparent governance and constructive engagement with those living and working nearby.

International experience increasingly suggests that these dimensions cannot be considered separately. The long-term competitiveness of the industry is closely linked to public confidence, environmental stewardship and the ability of communities and developers to understand each other’s perspectives.

Malaysia therefore has a unique opportunity. As one of the newer major data centre destinations, it can benefit from the technical innovations and policy lessons developed elsewhere while shaping its own approach to balancing economic development, environmental responsibility and community wellbeing.

MDCO Insight: Malaysia’s emergence as a regional data centre hub presents an opportunity not only to host global digital infrastructure, but also to demonstrate how such infrastructure can develop responsibly alongside local communities.

The Observatory Perspective

A data centre may appear, at first glance, to be simply another large industrial building. Behind its walls, however, lies an intricate interaction of engineering systems, digital technologies, financial investment, public policy, environmental management and human expectations.

Understanding data centres therefore requires looking beyond computers alone.

The engineering systems that keep digital services running are shaped by international technical standards, commercial requirements and society’s growing expectation that digital services should remain available continuously. The environmental and community impacts associated with these facilities arise not because they are separate issues, but because they are connected to the same systems that deliver reliability, security and performance. Likewise, the opportunities they create—in economic development, technological capability and artificial intelligence—are closely linked to the physical infrastructure built within local communities.

The Malaysia Data Centre Observatory adopts this broader systems perspective.

Rather than viewing data centres solely as engines of economic growth or as sources of environmental concern, MDCO seeks to understand how the many interconnected systems surrounding them influence one another. Greater transparency and a shared understanding of these relationships can help reduce information asymmetry, improve public discourse and support better decision-making by governments, developers, investors, regulators, professionals and communities alike.

This article serves as the starting point for the MDCO Explain series. The articles that follow explore the infrastructure systems, market participants, regulatory frameworks, approval processes and international experiences that collectively shape the rapidly evolving data centre landscape. Together, they aim to help readers understand not only what a data centre is, but why it has become one of the defining pieces of infrastructure of the digital age.

Selected References

1. International Energy Agency (IEA). Energy and AI (2025). https://www.iea.org/reports/energy-and-ai
2. International Energy Agency (IEA). Data Centres and Data Transmission Networks. https://www.iea.org/energy-system/buildings/data-centres-and-data-transmission-networks
3. Uptime Institute. Annual Global Data Center Survey (latest edition). https://uptimeinstitute.com/resources/research-and-reports
4. International Organization for Standardization (ISO). ISO/IEC 22237 — Information Technology — Data Centre Facilities and Infrastructures. https://www.iso.org
5. ASHRAE Technical Committee 9.9. Thermal Guidelines for Data Processing Environments. https://tc0909.ashraetcs.org
6. Synergy Research Group. Hyperscale Data Center Market. https://www.srgresearch.com
7. Malaysia Digital Economy Corporation (MDEC). Digital Infrastructure and Data Centre Initiatives. https://mdec.my
8. Malaysian Investment Development Authority (MIDA). Data Centre Investment Information and Announcements. https://www.mida.gov.my

Citation

Malaysia Data Centre Observatory (MDCO). What Is a Data Centre? MDCO Explain Series No. E01 (Version 1.0, July 2026).

MDCO Note

This article forms part of the Malaysia Data Centre Observatory (MDCO) Explain Series, which seeks to improve understanding of Malaysia’s data centre ecosystem through independent, evidence-based and balanced analysis. It is intended for educational and informational purposes only and does not constitute legal, engineering, planning, environmental, financial or other professional advice.

Malaysia’s rapidly evolving data centre ecosystem includes facilities developed, owned or operated by organisations such as AirTrunk, Amazon Web Services (AWS), Bridge Data Centres, DayOne, EdgeConneX, Google, K2 Data Centres, Microsoft, NTT Global Data Centers, Princeton Digital Group (PDG), ST Telemedia Global Data Centres (STT GDC), STACK Infrastructure, Vantage Data Centers, YTL Data Centre Park and many others. MDCO is independent of these organisations, as well as governments, regulators, utilities and advocacy groups. Its role is to facilitate transparency, structured understanding and equal access to information by presenting publicly verifiable evidence, relevant context and multiple stakeholder perspectives. MDCO does not endorse, oppose or advocate for any particular organisation, project or policy position.