We provide users with a Technical Exchange platform where we offer solutions to common issues encountered during daily product use. If you have additional questions, please reach out to us through the Contact Us section, and we will do our best to address any concerns.

Most applications and cloud services rely on data centers, but what exactly makes up these core systems of the online world? Global interconnection and data center company Equinix last week unveiled its new data center in Slough, England (built at a cost of 145 million dollars). This multi-story glass and steel building will initially house thousands of servers in a 236,000-square-foot hall.
Over the past 15 years, Equinix has invested a cumulative total of over 7 billion dollars in data centers that make up its International Business Exchange system, and has recently built new data centers in New York, Toronto, Melbourne, Shanghai, and Singapore. The LD6 data center is the sixth Equinix data center in the London area and one of over 100 data centers it operates across 33 markets worldwide.

LD6 can initially accommodate 1,385 server cabinets in its 236,000-square-foot data hall, with room for an additional 1,385 cabinets in the future.
The hall will house co-located servers, with each customer typically deploying their own machines. Servers requiring high-security environments will be placed in dedicated cages.
Equinix did not disclose specific customers signed up for the data center, but indicated that financial services, network services, and cloud service providers have expressed interest.
To maximize the available space for computing infrastructure, the building was designed to require support columns only in the center.
The hall supports up to 1.5 kilowatts per square meter for equipment and 12 kVA of cooling capacity per cabinet.

The data center is monitored around the clock and features biometric access controls and mantraps to prevent unauthorized access. A mantrap is a small room with an entry door on one wall and an exit door on the opposite wall. One door of the mantrap can only be unlocked or opened when the opposite door is closed or locked.
The building management system scans the infrastructure every 15 seconds, taking temperature and humidity readings, and reports the results to the on-site monitoring center.
The building was specifically designed to provide optimal duct width for the fresh-air cooling system.
Russell Poole, Equinix's UK Managing Director, said: "We actually started with the cooling technology, and then designed the building around it."
It took approximately two and a half years to fill the colocation space provided by LD6 at another Equinix data center.

Data cables are routed overhead, while power and cooling are delivered from below the floor.
Fiber optic cross-connects are housed in yellow cable trays, with copper cross-connects above, connecting down to the server cabinets below.

The core of the building is dedicated to a mechanical design that keeps servers running continuously and delivering data, with appropriate cooling measures in place.
The building uses fresh outside air rather than mechanically cooled air to cool the servers, minimizing energy consumption.
Air is drawn in from outside and passed through heat exchangers, which cool the hot air extracted from the server cabinets. This cooled air is then delivered back through floor vents to the data hall and server cabinets, and the cycle begins again. The system helps maintain server room temperatures at a maximum of 22 degrees Celsius, with no mixing or contamination of airflows.

The pipe running up the side of the building draws water from a borehole drilled 350 meters underground.
When temperatures exceed 20 degrees Celsius, water is sprayed onto the heat exchangers and evaporators to remove extra heat from the air circulating through the data center. If temperatures reach 30 degrees Celsius, a chilled water system is activated to dissipate additional heat.

By minimizing the energy required to cool the server halls, Equinix achieves a Power Usage Effectiveness (PUE) rating of just 1.2 when operating the building, ahead of the industry average of 1.7. PUE measures how much of a data center's total energy consumption goes toward running servers versus cooling and other infrastructure.
The LD6 data center has achieved LEED (Leadership in Energy and Environmental Design) Gold certification.

Each server is fed by two separate power supplies, so if a power outage occurs, one supply can always take over.
Each power line is protected by two uninterruptible power supply (UPS) units, capable of sustaining the data center at full load for up to 8 minutes.
These 8 minutes should give the data center's diesel generators enough time to start producing power. Once the data center is fully completed, 32 generators should be able to keep it running for 36 hours.
Power supplied to the building's cooling, fire suppression, and security systems is also protected by UPS.

The LD6 data center brings the number of Equinix data centers on the Slough campus to three, alongside the existing LD5 and LD4 buildings. The campus is connected to transatlantic cables and is one of the UK's busiest network nodes, offering latency of around 300 milliseconds to New York and approximately 4 milliseconds to Frankfurt.
Equinix's data centers house computing and network systems that serve a significant portion of the financial services industry's customers: a quarter of European securities trades originate from this Slough campus.

Once the second phase of the LD6 project is completed, the three data centers on the Slough campus will provide over 388,000 square feet (36,000 square meters) of colocation space, interconnected by more than 1,000 dark fiber links.
These links provide low-latency connectivity to over 100 network and cloud service providers located on the campus, including Microsoft Azure, Amazon Web Services (AWS), and Google Cloud. In addition, these high-bandwidth links enable customers to distribute their IT infrastructure across multiple locations, providing greater resilience against outages and allowing them to rapidly provision additional computing capacity when needed.

Source: 51CTO.com