Over the past few years, almost every enterprise, and most individuals as well, have migrated to the cloud, or are in the process of migrating. This has completely changed the way the world does business.
Rather than setting things up from scratch, it makes more sense for most enterprises to go to a cloud provider that already has the requisite infrastructure and artificial intelligence (AI) specific infrastructure (such as graphics processing units (GPUs), tensor processing units and cybersecurity) in place. The cloud allows easy scaling of computing, data storage and processing, with analytics on demand. “Just-in-time” scaling vastly reduces the need for upfront investments. Cloud pay-per-consumption models consistently deliver 30-40 per cent savings compared to traditional infrastructure.
In addition, the integration of edge computing enables real-time data processing and low-latency applications. Most cloud service providers already offer some form of AI as-a-service, making other advanced tools accessible.
Thus, the cloud helps corporates to quickly upgrade IT infrastructure and digital capability at minimal cost. Inevitably, this has led to a situation where new technologies are developed to run natively on the cloud and, in turn, this creates a feedback loop where every user is moving to the cloud.
During Covid-19, India’s ministries adopted cloud-based disaster recovery frameworks. This resulted in documented performance gains such as Gujarat achieving up to 40 per cent faster service delivery, while also enhancing disaster recovery.
India’s 5G subscriber base has crossed 400 million, positioning the country as the world’s second-largest market for the technology, behind only China. The 5G roll-out is transforming relationships for cloud, edge and end-users, enabling new applications. 5G can offer support for immersive augmented and virtual reality, along with the management of billions of internet of things (IoT) devices.
Along with 5G deployment and IoT management, AI penetration is a big driver for cloud adoption. AI needs scalable, high-performance cloud environments, massive data storage, etc. The demand for resources to handle the massive workloads imposed by large language models and real-time inferencing drives cloud-based network infrastructures.
The projected growth rates are impressive. In 2024, global spending on cloud services was over $706 billion, and it would have reached $1.3 trillion by end 2025. The market for cloud-managed networks is projected to grow from $27.99 billion in 2025 to $63.58 billion by 2033, reflecting a CAGR of 10.8 per cent over the forecast period. Similarly, the multicloud networking market is forecast to expand from $4.2 billion in 2024 to over $32 billion by 2034, a CAGR of over 22 per cent. The broader cloud network infrastructure, which includes software-defined wide area networks (WANs), virtualised routing and orchestration platforms, is estimated to grow from $232 billion in 2023 to $686 billion by 2032.
Given India’s policy focus and massive digital scale, it will be at the forefront of cloud adoption across the next five years and beyond. The country already has one of the world’s largest digital public infrastructures, and its digital growth rates are far outpacing global trends.
Consider platforms such as Aadhaar, the Unified Payments Interface and DigiLocker. The National Informatics Centre operates over 100 petabytes of storage and over 5,000 servers supporting core e-governance platforms. Ensuring continuing economic and strategic autonomy in the digital domain, along with sustainability, will be an ongoing challenge.
Market dynamics
Popular modes of cloud deployment include infrastructure-as-a-service (IaaS), platform-as-a-service (PaaS) and software-as-a-service (SaaS). IaaS offers virtualised resources, allowing businesses to remotely rent servers, storage and networking capabilities, eliminating the need to maintain hardware. PaaS offers tools and frameworks for developers to build and deploy applications. SaaS delivers fully functional applications, without users being responsible for installation and maintenance.
The commercial adoption of the cloud varies widely, since priorities differ with digital maturity. Start-ups with zero legacy work on the cloud from day one, using SaaS and PaaS to achieve scale without capex tied to physical infrastructure. Meanwhile, established enterprises must manage combinations of private cloud, public cloud and edge due to legacy.
Organisations may choose from a mix of on-premises, multicloud and edge solutions. On-premises cloud computing (also called private cloud computing) deploys infrastructure within the organisation’s own data centres. This ensures control, data residency compliance and security, and is preferred by government organisations, financial institutions and heavily regulated sectors.
In multicloud models, clients shop around across service providers to find the most suitable mix for their needs in terms of cost, performance and compliance. Hence, they use services from several providers, reducing single-vendor dependency while ensuring redundancy and disaster recovery. This helps optimise costs as well. Edge computing extends cloud capabilities closer to data sources at low latency to enable real-time analytics. This is ideal for IoT-driven applications, autonomous systems and remote industrial sites.
Going green
Cloud computing is extremely energy-intensive. Data centres require substantial amounts of power to run and cool servers. Thus, it has a significant carbon footprint, with data centres contributing up to 3.7 per cent of global carbon emissions.
Sustainability and energy efficiency are imperatives and must be baked into data centre design. Green cloud technologies use renewable energy to reduce the carbon footprint by, for example, deploying virtualisation to reduce the need for physical servers.
Telco cloud models
Cloud adoption has altered business models for telecom service providers, infrastructure providers, equipment vendors and software developers. It enables a shift from hardware-heavy models to software-defined networks, allowing telcos to launch new services quickly and respond flexibly to customer needs while lowering costs.
As telcos have integrated cloud-native architectures, they have been able to scale networks dynamically, optimising traffic routing, and offer enterprise services such as network slicing and edge computing. This has helped create new revenue streams.
This is a dynamic, rapidly evolving environment. It will continue to open up new opportunities for telcos, cloud providers, software and AI platform developers. Software providers now deliver virtualised network functions, orchestration tools and cloud security solutions, and run AI-powered network management platforms.
Infrastructure providers and vendors have also discovered new opportunities. As demand has risen for data centres, cloud exchange points and fibre networks, there is a need for upgrading physical infrastructure.
Challenges and stumbling blocks
One of the biggest concerns is ensuring data security and privacy. The cloud is inherently more open to data breaches, unauthorised access and cyberattacks. Investment and oversight are required to ensure end-to-end encryption, strong authentication protocols and stringent regulatory standards.
According to the latest report by CloudSEK, India is the second-most targeted nation after the US in terms of cyberattacks. At least 95 Indian entities suffered data theft attacks in 2024. One interesting set of cybersecurity solutions involves using blockchain-as-a-service to secure multi-party data sharing and create transparent audit trails. However, blockchains face challenges of scalability, energy efficiency and regulatory uncertainty.
Other challenges pertain to latency and reliability. Real-time applications such as videoconferencing, IoT or telemedicine require stable, high speed connectivity, and poor network coverage hinders cloud services. While new concepts such as satellite broadband and satellite cloud may improve network coverage, these may also lead to new technological issues, and the costs of satellite cloud could alter commercial equations.
In practical terms, enterprises are increasingly discovering that vendor lock-in can be a concern. Transitioning from one cloud service provider to another can be challenging due to proprietary architectures, data transfer limitations or a lack of interoperability.
Interoperability between multicloud and hybrid environments is another issue. A mix of public, private and on-premises solutions is common. However, a lack of seamless integration leads to inefficiencies and increases management complexity. Managing compliance across jurisdictions is another significant challenge. This is where the sovereign cloud enters the picture.
Going beyond the technical aspects, understanding app dependencies, evaluating on-premises versus cloud costs and assessing technical feasibility are among the biggest cloud migration challenges. Notwithstanding all the genuine issues and challenges, cloud migration is inevitable and accelerating.
Given the rise of remote data storage and the associated security challenges, new regulatory frameworks are needed. Regulation often lags behind technological change and the pace of change in this sphere is unprecedented. Nevertheless, regulatory frameworks are evolving. Cloud connectivity solutions must not only balance performance and sustainability but also be compliant with regulations in different jurisdictions. This has led to the concept of the sovereign cloud, in which data is managed in compliance with local regulatory environments.
Telecommunications networks have been early cloud adopters and moved from physical and virtual network functions to cloud-native network functions (CNFs). CNFs are installed in containers within an architecture of microservices. Containerisation packages software (and the dependencies) in isolated units, while microservices are sets of small, autonomous software services. Another common cloud-native technology is dynamic orchestration, automating the administration and coordination of services. CNFs allow telcos to offer services such as real-time analytics, immersive media streaming and IoT, while reducing operational complexity and cost.
The Indian cloud-native applications market had estimated revenues of $351.4 million in 2024, and may reach $1,593.6 million by 2030. That is an expected CAGR of 29.1 per cent from 2025 to 2030. Platforms deployed through private and hybrid clouds were among the top revenue-generating components in 2024, and the services segment deployed on public cloud is expected to grow fastest between 2025 and 2030.
Enterprises are distributing workloads across private clouds, public clouds and edge environments. This allows them the freedom to choose the best platform while reducing single-vendor dependence and ensuring enhanced compliance and data sovereignty.
In private clouds, all resources are dedicated exclusively to and accessible only by a single customer. Industry estimates indicate that India’s private cloud market could reach revenues of $35,235.5 million by 2030, at a CAGR of 16.2 per cent.
Hybrid and multicloud strategies help to balance cost, control and performance. India’s hybrid cloud market is projected to expand at a CAGR of over 19.29 per cent between 2025 and 2030. The multicloud management market was worth $156.8 million in 2022 and may reach $1,498.5 million by 2030, at a CAGR of 32.6 per cent. Cloud automation was the top revenue generator in 2022, but security and risk management may have the best growth prospects. It may be noted that every Indian market is expected to grow appreciably faster than the global market.
The rise of AI
AI dynamically allocates resources and manages workloads, making it indispensable for cloud infrastructure. Tools such as Kubernetes can create scalable infrastructure to handle large workloads. Automated machine learning (ML) and generative AI models deployed on edge devices aid with real-time decision-making. India’s cloud AI market generated revenues of $6,914.4 million in 2025 and is projected to reach $185,949 million by 2033, expanding at a CAGR of 50.5 per cent over the 2026-2033 forecast period.
AI is also being integrated into cloud WANs and GPU-optimised services. Cloud WAN reduces costs and delivers high performance securely across diverse geographies. Cloud GPUs may enable Managed Kubernetes Services, AI and ML solutions in an energy-efficient manner. GPU PaaS brings AI inference to the edge.
Serverless computing
Serverless computing is another trend that is redefining how developers deploy applications. Serverless computing abstracts underlying infrastructure, allowing developers to focus on writing and deploying code while cloud providers manage provisioning, scaling and maintenance of servers. This is normally done via pay-per-use. Serverless computing may soon become a default approach. However, the drawbacks include limited execution time, security concerns and complicated debugging.
India’s serverless computing market generated revenues of $1.3 million in 2024 and is expected to reach $3.4 million by 2030, at a 17.9 per cent CAGR between 2025 and 2030. In terms of segments, function-as-a-service was the largest revenue-generating model in 2024.
In the future, along with satellite cloud, we may see new technologies such as quantum cloud computing, which would revolutionise cryptography and solve currently intractable computational tasks. However, this is still very much at the experimental and proof-of-concept stage.
Sovereign cloud
Given India’s extensive public digital infrastructure and the momentum in digital connectivity, data security and sovereignty have become critical concerns. Regulatory compliance is particularly challenging, as it extends beyond data localisation.
Sovereign cloud guarantees data, operational and legal control over nationally significant data, which is non-negotiable. Countries such as Germany, France and Singapore have already partnered with global technology providers to build sovereign clouds, and India needs to emulate them. China’s cloud ecosystem has three national infrastructure providers. Europe blends sovereign controls with hyperscale technology.
India generates nearly 20 per cent of global data and must maintain jurisdictional control over its digital infrastructure. Today, over 300 departments across the centre and states use cloud services. But much of India’s digital financial transactions, health records and AI models trained on Indian data are hosted on infrastructure governed by foreign legal regimes. Much data that is physically stored in India remains subject to foreign subpoenas and extraterritorial laws. This is not acceptable. The Digital Personal Data Protection (DPDP) Act, 2023 was tightened under the DPDP Rules, 2025, effective November 13, 2025. Requirements include purpose-bound access, mandatory controls, strict residency and near-instant breach reporting. These must be incorporated into the sovereign cloud framework, guaranteeing control over data, digital infrastructure, operational authority and legal jurisdiction.
The government must act as an anchor tenant, migrating foundational platforms such as Aadhaar, GSTN and key e-governance systems to sovereign clouds. Critical sectors, such as government, defence, power, healthcare and banking, financial services and insurance, must mandate the use of sovereign cloud.
Conclusion
The cloud computing landscape is dynamic, interconnected and evolving at a rapid pace. The cloud is already foundational to innovation, enabling businesses to adapt, remain competitive, improve efficiency and unlock new revenue streams.
Devangshu Datta
