Feature
Public cloud computing has been growing uninterrupted over the past 10 years, with no signs of abating. Gartner is expecting investment in public cloud service to increase more than 20% in 2024, to $679 billion, up from $564 billion in 2023.
Anyone keeping up with public cloud investment figures seems to agree. Recent data from IDC shows worldwide public cloud services revenue totaled more than $315 billion in the first half of 2023, an increase of 19% over the same period in 2022. Statista also predicted that the public cloud market would keep growing between 2023 and 2028, by $466 billion, which constitutes growth of 78% to a value of $1.1 trillion by 2028.
The picture for hybrid cloud computing — an approach that combines at least one private cloud with at least one public cloud — is also bright. In 2021, the global hybrid cloud market was valued at $85 billion, according to researchers at Market Research Future. It is now predicted to reach $320 billion by 2032.
The private cloud sector is also estimated to grow, at a rate of 27% per annum between 2023 and 2028, with the market size forecasted to increase by $619 billion.
So, what is driving all this growth? We took a closer look.
Generative AI Drives Cloud Growth
According to Gartner's release, the rise of generative AI will be driving a large part of the growth in cloud investment.
"The tables are turning for cloud providers, as cloud models no longer drive business outcomes but rather, business outcomes shape cloud models,” wrote Sid Nag, vice president analyst at Gartner, in the company's press release. “Organizations deploying generative AI services will look to the public cloud, given the scale of the infrastructure required.”
Mayank Jindal, software development engineer at Amazon, agrees GenAI will be an impetus for cloud investment. But, he said, this isn't all because of the technical requirements of GenAI deployments. Rather, he believes a big part of that growth will stem from companies addressing the non-technical issues of GenAI, such as cost efficiency, economic factors, data sovereignty, privacy concerns and sustainability. Because of that, the focus will be on using the vast capabilities of hyperscale cloud service providers to support the heavy computational demands of generative AI applications.
He forecasts investments will be even larger in the public cloud arena, as companies seek more scalable, powerful computing resources to deploy generative AI services.
The public cloud offers a level of scalability, flexibility and efficiency that is challenging to achieve with private or on-premises solutions. Additionally, addressing non-technical aspects like privacy and sustainability are crucial for organizations to maintain regulatory compliance and corporate responsibility while harnessing the benefits of advanced AI technologies. Private clouds or on-premises data centers, while offering better control over data and operations, often come with higher costs and limited scalability, which aren't likely to meet the high computing demands required to operate generative AI services, Jindal said.
Related Article: Is Your Cloud Service Provider Trustworthy?
4 Trending Public Cloud Investments
While the kinds of public cloud services organizations have been investing in over the years have changed right now there are four principal trends, according to CIO Services CEO Eric Egolf. These include:
1. Hosting Server Instances
Hosting server instances moves what used to be stored in a datacenter — or closet in the case of a small business — to the public cloud, he said.
Organizations are still responsible for managing and maintaining those server instances either directly or through a service provider, he continued. Companies choose to move static workloads to the cloud to get out of the business operating hardware or due to the flexibility the public cloud brings.
For more dynamic workloads, which might be characterized as having scale up/scale down requirements by three times or more (Max Requirements Compute Requirements divided by Min Computer Requirements), the public cloud is still seen as an effective way to achieve that scale in a cost effective manner.
2. Platform as a Service – App Building Blocks
PaaS enables organization to get the same application-level results without having to manage server instances. AWS dynamo and Azure Dataverse are examples of database as a service in which no server instances are needed.
These services are building blocks for application development that may or may not also have server instance hosting, said Egolf. The reason for moving to these services is to increase application development time and decrease complexity and cost to maintain the application, he continued. It is much easier to connect your code to a database in a PaaS than spin up a SQL or mySQL server and secure and manage it.
3. Modern Applications – Public Cloud Services
Organizations are adopting modern applications like Teams and anything else in the Microsoft 365 or G Suite ecosystems. They are also purchasing more browser-based software as a service (SaaS) applications, he said.
Whether the SaaS application is hosted on a public cloud or private cloud is all the same to the customer, Egolf explained. Organizations are after flexibility, stability and ease of deliverability of applications to users. These applications, if they are feature comparable, with traditional client server applications are seen as better investments for the company.
Conference
Sep
16
HR Tech Conference Las Vegas 2025
Register
Conference
Oct
27
Gartner HR Symposium/Xpo Orlando 2025
Register
4. AI – Generative AI
AI and generative AI, he said, is a new world and will eventually fall under the Platform-as-a-Service umbrella. But for now, most companies are either taking a wait and see or in an experimentation phase.
The Gen AI ecosystem is still dominated by python language, although the tools in the ecosystem expand every day. Many of Egolf's customers are experimenting with RAG (Retrieval Augmented Generation) techniques using these types of platforms as they do not need to host the LLM itself.
“Although somewhat speculative I believe customers are spending money here because they don’t even know enough about their long-term needs and know Gen AI is moving so quick that making capital investments outside public cloud would be like playing darts blind folded,” he said. "If a business is going to make capital investments in their datacenter rather than public cloud, they would want to be able to depreciate the hardware and licenses over a minimum of four years."