The growth of the data centre business in India will be primarily driven by companies working on artificial intelligence (AI) and data inference rather than developing new large or small language models, Rajesh Tapadia, the chief executive officer of Iron Mountain’s data centres India business, said.
From an AI workload perspective, though both global hyperscalers and Indian startups have begun treating India as the new training ground for large language models (LLMs), the next phase of growth will continue to be driven by AI inference companies, Tapadia added.
“Globally, we do not create Gigawatt-scale campuses. Our sweet spot is creating 100-500 Megawatt (Mw)-capacity data centres, and that is where we are primarily positioning ourselves. In India, the inference workload growth will be much higher,” he said, adding that since the company’s campuses will have enough capacity for both, Iron Mountain will be able to cater to some of the domestic demand from Indian startups working towards LLM training.
The company is looking to add another 150 Mw of data centre capacity in India over the next three years, with three campuses that are currently under construction in Mumbai, Chennai and Noida. These campuses have planned capacities of 85 Mw, 50 Mw and 15 Mw, respectively and will go live by 2028, he said.
With increasing AI workloads, the demand from data centres is also changing, Tapadia said, adding that most of the legacy data centres are now being retrofitted with new capacity servers, racks, and cooling infrastructure to handle the increased workload.
“We do not have that much of a legacy in India, but that is an advantage for us. All of our buildings are new and AI-ready, which primarily means that, depending on the construction of the facility, we can quickly convert the entire facility into an AI-native facility,” Tapadia said.
One of the other benefits of the data centre business in India is the surplus availability of stable power, along with green power generated from renewable energy sources such as wind, solar, and hydro, he said.
“India has all the ingredients to be a regional hub for data centres, and there are multiple drivers for it. Obviously, the domestic market itself is booming with rapid digitalisation and increased cloud adoption. AI is accelerating the overall demand,” Tapadia said, adding that tax incentives such as the 20-year tax break announced during the Budget are the right policy interventions designed to take the data centre growth in India to the next level.
 "AI inferencing to drive India data centre growth: Iron Mountain's Tapadia")
