Indian IT giant Tata Consultancy Services (TCS) has partnered with private equity firm TPG on a multi-year, $2 billion project to establish a network of gigawatt-scale AI data centers across India. This ambitious initiative, dubbed "HyperVault," aims to address the rapidly escalating demand for AI compute power, which is currently outpacing the development of essential infrastructure. TPG is contributing $1 billion to the venture.
India faces a significant disparity in AI compute capacity; while it generates nearly 20% of the world's data, it accounts for only about 3% of global data center capacity. This gap has prompted major tech companies and cloud providers to invest billions in expanding local infrastructure, capitalizing on India's increasing adoption of AI products.
The HyperVault project will develop liquid-cooled, high-density data centers designed to support advanced AI workloads across key cloud regions. While liquid cooling and high-density rack designs are becoming standard for power-hungry AI GPUs, they raise concerns in countries like India, where water scarcity is a pressing issue. Urban hubs such as Mumbai, Bengaluru, and Chennai, which concentrate much of India's existing data center capacity, already experience water stress. S&P Global, citing Uptime Institute estimates, notes that a 1-megawatt data center can require up to 25.5 million liters of water annually for cooling, further straining infrastructure. Beyond water, the rapid expansion of AI data centers also stresses India's power supply and land use – two other critical bottlenecks identified by industry analysts. High-density AI clusters demand reliable electricity and large industrial land parcels, resources that are increasingly difficult to secure in major urban areas.
Despite these challenges, global tech companies view India as a crucial frontier for building AI infrastructure. Over the past two years, local and international technology firms have announced investments exceeding $32 billion to expand data center infrastructure in the country, according to S&P Global. For instance, Microsoft committed $3 billion over two years for India's cloud and AI infrastructure in January. In October, Google announced plans to spend $15 billion over five years to build a gigawatt-scale AI data center hub in Andhra Pradesh. Amazon also pledged $12.7 billion by 2030 for AWS cloud infrastructure in India. Local players like Reliance Industries and CtrlS are also significantly expanding their data center capacities.
TCS stated its intention to collaborate with hyperscalers and AI companies to design, deploy, and operate this expanding AI infrastructure platform. The company aims to build approximately 1.2 gigawatts of capacity in its initial phase. S&P Global estimates that over 95% of India's new data center capacity in the next five years will come from leased facilities, with the remainder driven by hyperscalers constructing dedicated AI infrastructure. TCS and TPG project that India's total data center capacity could surpass 10 gigawatts by 2030, a substantial increase from its current approximate 1.5 gigawatts.
