India is witnessing a surge in digital connectivity, driven by 5G roll-outs, increasing internet of things penetration and the digitalisation of services. Smart homes, smart factories and connected cars are generating vast amounts of data, which is processed and stored in real time in data centres and runs on cloud-native platforms. As a result, the information and communications technology industry is facing increasing pressure to decarbonise its energy-intensive infrastructure. As per industry estimates, installed data centre capacity in India could grow from 1.5–2 GW in 2024-26 to around 17 GW by 2030. This expansion could account for nearly 8 per cent of India’s total electricity demand by the end of the decade. Given India’s continued reliance on non-renewable energy sources, data centres will become a significant contributor to carbon emissions.
The solution lies in adopting green cloud. According to an industry report, a global shift to green cloud could cut total IT emissions by 5.9 per cent. Green cloud is a model that blends cloud-native flexibility with environmental accountability. It enables energy optimisation at software and hardware levels, reducing dependence on grid power. This can be a good financial decision for telcos as it will help reduce their energy bills. A look at the key technology enablers of green cloud, adoption challenges, current deployments and the way forward…
Key enablers
Server virtualisation and containerisation
The foundation of green cloud is software-first infrastructure. Instead of dedicating an entire physical server to just one telecom function, the workloads can operate as virtual machines or containers on shared hardware. The containers allow dozens of services to share compute, power and memory. It also leads to power saving through shutdown.
IaaS and elastic cloud infrastructure
Cloud service models such as infrastructure as-a-service (IaaS) enable operators to dynamically scale resources up or down. They allow operators to spin up clusters during peak demand, and later migrate or consolidate into leaner set-ups, thereby distributing workloads across data centres. This elastic use of resources ensures that power-intensive hardware is operated only when needed, helping save both time and emissions.
AI-driven load balancing and scheduling
Artificial intelligence (AI) processes historical patterns such as time of day, location and grid load to predict which compute resources to use or shut down. Moreover, AI-driven controllers can throttle or shut specific network components. When integrated with cloud-native orchestration platforms, this creates a unified and intelligent framework that maximises energy efficiency.
Edge sites with on-site renewables
Reducing emissions is not limited to optimising central data centres; it also extends to the network edge. Edge nodes, such as base stations, edge points of presence and radio access network (RAN) nodes, consume significant power to handle the traffic. By integrating solar photovoltaic (PV) systems, battery energy storage or even small-scale wind turbines, operators can power these units with on-site renewable energy. Moreover, localised processing at the edge reduces backhaul data transport to centralised data centres, which is an energy-intensive operation. Processing close to the source using clean power offers dual efficiency gains–lower transport energy consumption and reduced emissions.
SDN/NFV and network slicing
Software-defined networking (SDN) and network functions virtualisation (NFV) make networks programmable and energy aware. SDNs can monitor energy usage across the network, re-route traffic and deactivate underutilised hardware nodes, optimising where, when and how data is transmitted to minimise environmental impact. Meanwhile, NFV replaces dedicated telecom hardware such as routers and firewalls with software-based network functions. These functions can be consolidated on to fewer servers and scaled up when demand spikes and down during idle hours. This enables higher server utilisation and load migration, allowing idle hardware to be turned off completely, reducing both power consumption and cooling requirements. Further, network slicing enables operators to create customised virtual networks, each tailored to specific needs. These slices can be assigned an energy budget, ensuring they consume only what is necessary. These slices can be spun up and down based on actual demand, thereby enhancing energy efficiency.
Liquid cooling and efficient hardware
Cooling is one of the largest non-computational energy expense in data centres and edge telecom infrastructure. Processes such as immersion cooling and direct-to-chip cooling can reduce reliance on massive heating, ventilation and air conditioning (HVAC) units and significantly improve power usage effectiveness with reduced thermal throttling. Furthermore, systems like UPSs, inverters and generators can operate inefficiently under lighter workloads. Modernising these systems, by adopting modular UPS or lithium-ion batteries, minimises energy loss during power conversion.
Green SLAs and digital twins
Green service-level agreements (SLAs) between service providers and their clients include carbon efficiency targets, which incentivise operators to adopt energy-efficient measures and invest in renewable-powered infrastructure. Meanwhile, digital twin, when integrated with AI-based orchestration platforms, enables the simulation of changes and the assessment of their impact on performance and energy use, without disrupting real-world operations.
Challenges in adoption
Green technologies often require significant capital investment, which can be challenging for industries with thin margins. Deploying energy-efficient hardware, renewable energy infrastructure and other sustainable technologies entails significant capital expenditure, and recovering the investment through energy savings can take years. Moreover, in a price-sensitive market such as India, balancing green investment against other needs, like spectrum payments or debt reduction, becomes more difficult.
There is also the paradox of 5G densification and rising energy usage. As per an industry report, 5G networks are 90 per cent more efficient in terms of kilowatt per data bit than 4G. However, the increased network density and traffic are expected to offset these savings, resulting in overall 5G energy consumption being four to five times higher than 4G. In dense areas, the proliferation of microcells per square kilometre leads to significant cumulative power demand.
While renewable energy is a crucial aspect of green cloud, the availability of clean energy sources can be limited in certain locations. Many remote towers lack reliable grid access, forcing operators to fall back on conventional energy sources to keep networks operational. The Ministry of Electronics and IT has itself acknowledged that green energy is not always firm or reliable. Additionally, several regulatory issues complicate the green energy transition. For instance, securing right of way for developing renewable infrastructure can be slow and inconsistent. Further, weak coordination between agencies, such as the telecom, power and environment ministries, can slow down progress.
Current deployments
Bharti Airtel has launched the “Green 5G” initiative, which leverages AI and ML to optimise energy consumption across its 4G/5G RAN, aiming to reduce carbon emissions by approximately 143,400 metric tonnes annually. The company has also solarised over 30,000 network sites, reducing both emissions and operational costs. Meanwhile, Reliance Industries is developing a massive 3 GW AI-ready data centre in Jamnagar, Gujarat, which will be powered entirely by green energy. The Adani Group is also planning to source renewable power from a new solar-wind hybrid project at Khavda, expected to begin operations in the third quarter of 2025. Furthermore, various hyperscalers have committed to running all operations on 100 per cent renewable energy by 2025.
The way forward
Looking ahead, India is not only striving for faster networks but also laying the foundation for a telecom future that is sustainable from the ground up. Initiatives such as the Bharat 6G Alliance are already prioritising sustainability as a core objective, recognising that the next generation of connectivity must also address the urgent climate crisis.
As the third-largest emitter of greenhouse gases globally, India contributes over 4 billion tonnes of carbon emissions, which account for about 7 per cent of the global total. The country has committed to reaching net-zero emissions by 2070 and installing 500 GW of non-fossil fuel capacity by 2030, with renewables making up half of its total energy consumption. However, current realities tell a different story. Coal still powers over 68 per cent of the national grid and remains the largest single source of India’s emissions. Bridging this gap between ambition and execution requires a sector-wide transformation.
It is not just about running telecom operations on the cloud, but also about doing so in the greenest, smartest way possible. As India accelerates toward a digital-first economy, a climate-aligned telecom ecosystem will be critical to ensuring sustainable growth.
