Illustration of a server rack on a testing platform, symbolising AI infrastructure validation
Use promo code “TCH15” to get 15% off on checkout.

Digital Realty has opened a facility in Singapore where companies across Asia Pacific can test AI infrastructure before committing to it at scale, as the gap between AI ambition and the hardware to run it keeps growing in the region.

Key Takeaways
  • Digital Realty’s new Innovation Lab (DRIL) in Singapore sits inside its SIN11 facility, joining existing labs in Northern Virginia, Tokyo, London and São Paulo.
  • Its racks are liquid-cooled and built for cabinet densities of 150 kilowatts and above.
  • Half of IT decision-makers in Singapore now cite a lack of specialised infrastructure as their top obstacle to AI plans, up from a global 9% in 2024.
From left, Mr. Herman Loh, Co-Head, Digital Industry Singapore, Singapore Economic Development Board (EDB), Chris Sharp, Chief Technology Officer, Digital Realty, alongside Ms Aileen Chia, Deputy Chief Executive Officer, Infocomm Media Development Authority (IMDA), followed by Serene Nah, Managing Director and Head of Asia Pacific, Digital Realty, Ms. Cindy Koh, Group Director, Green Growth, JTC Corporation (JTC), and Govind Choudhary, General Manager for Southeast Asia and India, Digital Realty. 

A live data centre to test AI infrastructure

The Digital Realty Innovation Lab (DRIL) sits inside the company’s SIN11 facility in Singapore. It gives enterprises a working data centre environment to test how different vendors’ AI hardware, cooling and networking gear perform together, rather than finding out after it’s already deployed. The Singapore lab joins existing DRIL sites in Northern Virginia, Tokyo, London and, most recently, São Paulo.

Racks inside the lab are liquid-cooled and built to handle cabinet densities of 150 kilowatts and above, roughly what’s needed to run the GPU clusters behind large AI models.

“Enterprises deploying AI want to understand how different components and technologies can perform together before committing to infrastructure at scale,” said Chris Sharp, Chief Technology Officer at Digital Realty. “The DRIL will provide a production-grade testing environment where customers can work with leading technology partners and our engineering teams to validate AI, high-performance computing and hybrid cloud architectures under real-world conditions.”

Singapore firms more worried about AI infrastructure than most

Digital Realty’s own research gives some sense of why it built the lab here. In its 2026 Global Data Insights Survey, 40% of IT decision-makers worldwide named a lack of specialised infrastructure as the main thing holding back their AI plans, up sharply from just 9% in 2024. In Singapore, that figure is 50%.

Despite the bottleneck, nearly six in ten organisations across Asia Pacific, 59%, plan to raise their AI spending by more than a quarter in 2026.

Part of a wider Singapore build-out

The lab adds to Digital Realty’s existing investment in Singapore. In August, the company was selected under Singapore’s second Data Centre Call for Application to build a 50-megawatt AI-ready data centre on Jurong Island.

“Singapore’s strong regional connectivity makes it a strategic home for the DRIL, where enterprises in the rest of Asia Pacific can test their solutions with our partners in real time before deploying at scale,” said Serene Nah, Managing Director and Head of Asia Pacific at Digital Realty. “Together with our planned data centre facility on Jurong Island, the DRIL will help customers in Singapore and across the region move from concept to confidence.”

The lab’s opening was attended by officials from Singapore’s Infocomm Media Development Authority, the Economic Development Board and JTC Corporation, alongside Digital Realty executives and partners including HPE and Cisco, which are supplying some of the technology running inside the facility.

Author


Discover more from techcoffeehouse.com

Subscribe to get the latest posts sent to your email.

Discover more from techcoffeehouse.com

Subscribe now to keep reading and get access to the full archive.

Continue reading