Direct Liquid Cooling in Data Centres

What Is an Edge Data Centre?

Businesses today create, manage, and deliver more data than ever. But with users and devices located across many different regions, transferring all that data through central servers can increase latency, create bottlenecks, and decrease user satisfaction.

A key method of improving data delivery is by using edge data centres. Edge data centres are strategically located facilities designed to bring data storage and processing closer to end users. They enable quicker responses, improved reliability, and more efficient utilisation of network resources and computing infrastructure.

More than that, they ease the burden on those critical central data centres. As such, edge data centres have become essential for modern distributed computing environments, where milliseconds matter when transferring large amounts of data. This is reflected in recent estimates1, predicting that 75 per cent of data will be created in edge locations by 2025.

With demanding new technologies and latency-sensitive workloads growing, edge data centres are more impactful than ever. In this blog, we’ll see how they work, the different types, and industry use cases.

What is edge computing in simple terms?

Edge computing is the method of using strategically located computing resources to process data closer to end users. Being physically closer means edge data centres can deliver data with minimal latency.

Edge computing is especially crucial for services that demand rapid content delivery. For example, streaming services, high-traffic sites, and cloud computing operations can all benefit from serving data from edge locations.

Likewise, many cutting-edge technologies require massive computing power spread over several sites. Artificial intelligence is reshaping every industry, while Internet of Things (IoT) devices open up possibilities to generate data almost anywhere. These technologies can all use edge computing to improve data transfer efficiency.

How does an edge data centre work?

Edge data centres act as a connection between networks, other edge locations and central data centres. They are strategically positioned closer to data request sources — whether this means data processing, data creation, or end users. By processing closer to the source, they reduce data delivery time and deliver efficiently.

To achieve this, edge data centres employ edge caching. This involves receiving data from central servers and temporarily storing it to deliver when requested. This not only improves computing response times but also reduces demand on the overall network.

Using a network of edge data centres allows for much quicker access to data and resources — vital for functions like edge computing.

What is the difference between core and edge data centres?

Compared with edge data centres, core data centres are larger, traditional data centres that might act as the central hub for data storage. They may also be dedicated to highly sensitive or security-conscious information.

Edge data centres are smaller facilities dedicated to delivering data within specific regions. Increasingly, businesses use edge data centres to cache and deliver data received from central locations to improve the efficiency of transfers to end users.

Characteristics of edge data centres

An organisation or service might employ many edge data centres within its network. These edge data centres will all share some common characteristics:

Size: Edge facilities are typically smaller than traditional data centres. However, they are usually just as well-equipped, providing the same high-performance data storage and networking capabilities.

Location: Edge data centres are geographically closer to users, devices, and network carriers. Organisations can strategically select the locations closest to their workloads or users.

Data: Edge data centres deliver time-sensitive data faster by placing data storage closer to users. Without using edge data centres, central data centres can be tasked with providing data to users extremely far away, resulting in higher costs and increased latency.

Deployment: Edge data centres don’t act alone. They typically belong to a complex network of other edge and central data centres.

Digital Realty's metro data centre services — allows worldwide connectivity and low-latency delivery to all users, regardless of location.

What are the benefits of edge data centres?

Using edge data centres can bring many advantages, including:

Reduced latency

Fast data delivery means a better customer experience and high-speed transfers of latency-sensitive tasks.

Improved performance

Delivering data from the edge ensures workloads perform at their best, especially if large volumes of data are moved across a wide area.

Enhanced reliability

Edge data centres spread workloads across multiple sites, bringing increased reliability, improved uptime and disaster recovery options.

Scalability

Using cloud services means new resources can be deployed quickly, with load balancing across different sites to manage peaks in demand.

Sustainability

Using edge data centres can help organisations reduce network traffic and open up more options for ESG initiatives.

The types of edge data centres

In general, there are three primary types of edge data centres:

On-premises (establishment-based) edge data centres

These edge data centres are located on a customer's site. This allows for the lowest possible latency, high-security controls and complete oversight over computing infrastructure. However, they require extensive funds and expertise to set up and manage.

Network (mobile) edge data centres

These are typically used by network service providers to improve the performance of their networks. The uptake in 5G technologies has helped these centres send data at high speeds.

Regional edge data centres

The most common type, regional edge data centres are located in non-major geographic locations. They are typically connected to central data centres and used by cloud providers to offer services in that specific region, delivering data with low latency and high reliability.

Digital Realty's PlatformDIGITAL facilitates low-latency connections, allowing for the fast delivery of critical workloads.

Edge data centre use cases

With so many potential benefits, edge data centres are becoming common within several major industries.

Media & Entertainment

Rapid data transfers are essential to many of today’s high-bandwidth data processing services. Edge data centres support low-latency applications like streaming services, gaming, and real-time content delivery, improving the user experience for customers.

Healthcare

Medical IoT devices enable remote patient monitoring, real-time analysis of medical data, and telemedicine services. They use edge computing to provide low- latency monitoring, enhancing patient care.

Financial services

In sectors where milliseconds make a difference — like financial services — edge data centres can give a substantial competitive advantage. Trading can be completed faster with lower latency, helping high-volume banking companies ensure a better return-on-investment (ROI).

Manufacturing

Edge computing can enhance efficiency and safety in manufacturing. Using connected devices, it enables the real-time monitoring of equipment, predictive maintenance, and process automation, all resulting in smoother operations and cost savings.

Transportation

A recent Forbes report2 highlights how self-driving cars make several life-or-death decisions each second. Edge services support autonomous vehicles by processing sensor data locally, enabling faster decision-making.

Energy

Edge computing facilitates smart grid management in the energy sector. It brings new energy optimisation and predictive maintenance opportunities, improving the reliability, efficiency and ROI of energy systems.

Smart cities

Edge data centres support many innovative city applications. Initiatives like traffic management, public safety, environmental monitoring, and infrastructure optimisation can all use edge processing to provide more sustainable and efficient urban environments.

Deliver data at the edge

Edge data centres work within an overall network infrastructure to enhance data transfers. By delivering content closer to users, they ensure the low-latency delivery of critical workloads. This results in increased performance, reliability, cost-effectiveness, and user satisfaction.

Digital Realty’s carrier-neutral data centres can act as both core and edge locations. With a wide range of global data centres in key strategic locations, our connectivity offerings — Metro Connect and ServiceFabric — allow you to place your critical services for optimal performance.

Frequently asked questions (FAQs) about edge data centres

What is the main goal of edge computing?

The main goal of edge computing is to improve the speed and efficiency of data transfers. By placing edge centres across many locations, organisations can deliver key data and workloads faster, reducing overall network strain and improving the user experience.

What is the difference between edge and colocation?

Colocation services allow companies to use data centre space to store and run their important workloads. While central data centres are often used for this purpose, companies can also use edge data centres to reach more geographically spread end users with low latency and high reliability.

What is the difference between edge and hyperscale?

A hyperscale data centre is a massive facility dedicated to the largest, most demanding business-critical workloads. According to Synergy Research Group3, there were around 900 hyperscale data centres in the world in 2023. Edge data centres are smaller facilities that handle less demanding but geographically sensitive tasks.

External sources

Edge computing market size: https://www.fortunebusinessinsights.com/edge-computing-market-103760

Forbes self-driving cars article: https://www.forbes.com/sites/naveenjoshi/2022/08/05/5-moral-dilemmas-that-self-driving-cars-face-today/?sh=4c109751630d

SRG hyperscale centres: https://www.srgresearch.com/articles/hyperscale-data-center-capacity-to-almost-triple-in-next-six-years-driven-by-ai