Research Finder
Find by Keyword
Agentic AI, Cloud Deployment Models, and Hypervisors: Predicting the Enterprise IT Landscape of 2025
Agentic AI, sustainable data centers, workload repatriation, sovereign clouds, and the evolving hypervisor landscape are reshaping enterprise IT.
Key Highlights
Agentic AI is poised to automate complex tasks and drive innovation across industries.
Sustainability is no longer optional for data centers; it's a business imperative.
Hybrid cloud strategies and workload repatriation are gaining traction as organizations seek greater control and cost optimization.
Data sovereignty concerns are fueling the rise of sovereign clouds.
The hypervisor landscape is evolving beyond VMware, with Kubernetes and OpenStack gaining momentum.
Analyst Take
2024 was a year of AI experimentation, with enterprises exploring its potential through pilots and proofs of concept. However, McKinsey & Company's survey of over 400 C-suite executives reveals that meaningful adoption is still in its early stages. Many companies conducted proofs of concept without measurable benefits (41%) or achieved broader adoption with minimal gains (27%). Only 23% saw modest productivity increases from AI pilots.
This highlights the challenge of translating AI investments into scaled, transformative impact. Enterprises face obstacles such as data readiness, pipeline complexity, talent shortages, and ROI measurement. This situation presents an opportunity for IT vendors to help bridge the gap between AI aspirations and tangible results.
Apple, Microsoft, Amazon, Google, NVIDIA, Meta, and Tesla, known as the “Magnificent Seven”, dominated 2024. These tech giants drove innovation and investment in AI, influencing cloud computing, generative AI, and hardware.
However, with Broadcom reaching a trillion-dollar valuation, the question arises: will 2025 be the year it joins the ranks, creating a "Great Eight"? Broadcom's extensive involvement in AI chips, networking, and software positions it as a potential contender. Will the reign of the Magnificent Seven continue, or is it time to make room for Broadcom?
Looking ahead to 2025, constant flux and significant transformation will characterize the enterprise IT landscape. Five key trends are poised to reshape the industry, requiring organizations to adapt their strategies and embrace new technologies to stay competitive.
The Rise of Agentic AI and Data Centricity:
Move over, generative AI. While tools like ChatGPT have captured the imagination, the real game-changer is agentic AI. This more evolved form of artificial intelligence goes beyond generating content; it can independently define goals, make decisions, and take action in complex environments. This shift is driven by the increasing availability of data, advancements in machine learning, and the relentless pursuit of automation and efficiency. Imagine AI agents that not only respond to customer inquiries but proactively resolve issues, personalize product recommendations, and even manage entire marketing campaigns. In software development, agentic AI will automate complex tasks like debugging and refactoring, freeing up developers to focus on innovation. Supply chains will be optimized through AI-driven demand prediction and automated negotiations. Agentic AI is poised to become a collaborative partner in driving business success.
Another vector in enterprise AI adoptions is whether to bring AI to the data, or data to the AI? This debate is becoming increasingly pivotal as enterprises grapple with growing volumes of data and the complexities of processing it efficiently. Traditionally, moving data to centralized AI models has been the norm, but this approach is now being challenged by the rise of sovereign clouds, hybrid deployments, and data localization requirements. Sovereign cloud solutions from providers like Oracle, which has embedded its Exadata platforms within hyperscale environments, illustrate the appeal of processing data closer to its source to address compliance and latency concerns.
On the other hand, AI-native solutions that ingest and process data across distributed environments, such as Snowflake and Databricks, are enabling centralized data pipelines that simplify enterprise workflows. The choice between the two approaches depends on use case specifics—whether low-latency, high-throughput inference or regulatory-compliant data governance takes priority. This debate is not merely technical; it underscores broader questions about enterprise control, flexibility, and long-term strategic alignment in the age of AI.
Sustainable Datacenter Operations:
The increasing energy demands of AI workloads are driving a parallel focus on data center innovation, where sustainability is no longer optional, but imperative. Traditional air-cooled servers struggle to handle the heat generated by high-density hardware like NVIDIA's H100 GPUs and custom silicon solutions. This, coupled with rising energy costs, stricter environmental regulations, and pressure from stakeholders, is fueling a surge in sustainable data center solutions.
Liquid cooling is emerging as a critical enabler, offering superior thermal management and energy efficiency compared to air cooling. Companies like Lenovo, with their ThinkSystem Neptune servers, are leading the charge in integrating liquid cooling systems. These systems optimize performance, reduce energy consumption, and enable higher density deployments, ultimately extending hardware lifespan and lowering costs.
Beyond liquid cooling, the industry is witnessing a shift towards:
Immersion cooling: Submerging servers in dielectric fluids for even more efficient heat dissipation.
AI-powered optimization: Using AI to dynamically adjust cooling systems, power distribution, and server utilization in real-time.
Renewable energy sources: Increasing reliance on solar and wind power to reduce carbon footprint.
As enterprises scale their AI deployments, balancing performance with sustainability becomes crucial. This makes liquid cooling and other innovative cooling technologies core components of next-generation data centers. Consequently, partnerships between server manufacturers and colocation providers to retrofit facilities for liquid cooling will gain momentum, addressing both the scalability and environmental concerns of AI infrastructure. Ultimately, companies are realizing that sustainability is not just good for the planet; it's good for business.
Workload Repatriation and Hybrid Cloud Strategies:
While cloud adoption continues, a counter-trend is emerging: workload repatriation. Organizations are bringing certain workloads back on-premises or adopting hybrid cloud models. This shift is driven by several factors. Cloud costs can be unpredictable, especially for data-intensive applications. Concerns around data security, privacy, and regulatory compliance are also pushing sensitive workloads back in-house. For applications demanding low latency or high bandwidth, on-premises infrastructure often delivers superior performance. Finally, organizations are seeking to avoid vendor lock-in by diversifying their infrastructure strategies. Hybrid cloud platforms and edge computing are enabling organizations to achieve the optimal balance between cloud flexibility and on-premises control.
The Rise of Sovereign Clouds:
While cloud adoption continues, a counter-trend is emerging: workload repatriation. Organizations are bringing certain workloads back on-premises or adopting hybrid cloud models. This shift is driven by several factors. Cloud costs can be unpredictable, especially for data-intensive applications. Concerns around data security, privacy, and regulatory compliance are also pushing sensitive workloads back in-house. For applications demanding low latency or high bandwidth, on-premises infrastructure often delivers superior performance. Finally, organizations are seeking to avoid vendor lock-in by diversifying their infrastructure strategies. Hybrid cloud platforms and edge computing are enabling organizations to achieve the optimal balance between cloud flexibility and on-premises control.
The Evolving Hypervisor Landscape: Beyond VMware
The IT landscape is undergoing a significant transformation, with both Kubernetes and the hypervisor landscape evolving rapidly. Kubernetes is maturing beyond container orchestration, becoming a foundational layer for platform engineering. This approach centralizes and standardizes developer workflows, crucial for building and scaling AI-enabled applications. Kubernetes-based platforms now offer tools for automation, observability, and multi-cloud deployment, reducing operational complexity and driving adoption across industries. Vendors like Red Hat, VMware, and SUSE are capitalizing on this by enhancing their Kubernetes offerings with AI-focused extensions. As Kubernetes becomes the backbone of modern application infrastructure, platform engineering allows organizations to focus on innovation.
Simultaneously, the hypervisor landscape is shifting, driven by Broadcom's acquisition of VMware and the rise of alternatives like Kubernetes and OpenStack. Concerns over potential price hikes and licensing changes with VMware are prompting organizations to re-evaluate their virtualization strategies. This creates opportunities for Kubernetes, with its focus on containerization and microservices, and OpenStack, with its comprehensive private cloud solutions. This shift towards open-source and container-based virtualization is fueled by the rise of cloud-native applications, the need for agility, and the desire to avoid vendor lock-in. Ultimately, this evolution fosters a more diverse and competitive market, offering organizations greater choice and flexibility.
Looking Ahead:
These five trends paint a picture of an IT landscape undergoing rapid transformation. The key trend that HyperFRAME is most eager to follow is the rise of agentic AI. Based on what HyperFRAME is observing, this technology has the potential to revolutionize everything from customer service and marketing to software development and supply chain management. Going forward, HyperFRAME will be looking closely at how different industries adopt and leverage agentic AI, and how this technology impacts productivity, innovation, and competitiveness. When taking in the market as a whole, the advancements in AI are truly remarkable.
The early-stage adoption of AI reflects both a challenge and an opportunity. Enterprises are eager to move from experimentation to execution, and the vendors that can guide them through this transition will define the next chapter of AI adoption.
Steven Dickens | CEO HyperFRAME Research
Regarded as a luminary at the intersection of technology and business transformation, Steven Dickens is the CEO and Principal Analyst at HyperFRAME Research.
Ranked consistently among the Top 10 Analysts by AR Insights and a contributor to Forbes, Steven's expert perspectives are sought after by tier one media outlets such as The Wall Street Journal and CNBC, and he is a regular on TV networks including the Schwab Network and Bloomberg.