Through 2026, more than one-third of enterprises will deploy cloud-native AI and data platforms in information architectures, a new ISG report says.
“AI workloads are highly variable and require large amounts of compute resources, making them ideal for the cloud environment”
The ISG Buyers Guide for Cloud-Native AI and Data Platforms, produced by ISG Software Research, projects that through 2026, more than one-third of enterprises will deploy cloud-native AI and data platforms in their information architectures. Most leading cloud providers offer both AI and data capabilities, while independent software providers offer AI and data platforms built to operate natively in and across cloud environments, the report says.
The report points out that AI requires large amounts of high-quality data that needs to be prepared, engineered and organized to feed and train AI models. The volumes of data necessary for accurate AI models place a significant demand on computing resources, which can often be best met with elastic cloud platforms. The cost of these systems can be significant, and any inefficiencies in the process can exacerbate the costs, the report says.
“AI workloads are highly variable and require large amounts of compute resources, making them ideal for the cloud environment,” said David Menninger, executive director, ISG Software Research. “Cloud, AI and data platforms should be evaluated together, rather than independently, to make the most informed decisions when selecting software providers.”
The increasing importance of intelligent operational applications driven by AI insights is blurring the lines that have traditionally divided the requirements for AI platforms and data platforms, the report says. While there have always been general-purpose databases that could be used for both analytic and operational workloads, traditional architectures have involved the extraction, transformation and loading of data from the operational data platform into an external analytic or AI platform. This enables the operational and analytic workloads to run concurrently without adversely impacting each other, protecting the performance of both.
The report notes the importance of coordinating cloud, AI and data efforts. Cloud platform providers have recognized the opportunity to help enterprises with this convergence, and all the top cloud providers are offering platforms that combine AI and data capabilities.
The Cloud-Native AI and Data Platform Buyers Guide includes an evaluation of platforms that provide three sets of capabilities: cloud, AI and data. To be considered for inclusion in this Buyers Guide, a product must offer services addressing key elements of cloud platforms that:
- Support a combination of public, private and hybrid cloud workloads;
- Include a general-purpose data platform, database, database management system, data warehouse, data lake or data lakehouse;
- Include data persistence, data management, data processing and data query functionality;
- Support database administrator, developer, data engineering and data architect functionality, and
- Support the AI-related capabilities of data preparation, AI/ML modeling, AutoML, GenAI, developer and data scientist tooling, MLOps/LLMOps, model deployment, model tuning and optimization.
