Cloudera has announced Cloudera Anywhere Cloud, a new platform aimed at bringing data applications and artificial intelligence into production on hybrid infrastructures. The solution allows deploying services across public clouds, sovereign infrastructures, and private data centers from a common control plane, with particular focus on a growing challenge in enterprise AI: using data where it’s stored without having to move it to a specific cloud.
The key points of Cloudera Anywhere Cloud in 30 seconds
- Cloudera Anywhere Cloud enables running data and AI services across public clouds, sovereign infrastructures, and private data centers.
- The platform aims to prevent data migrations for certain AI use cases while maintaining centralized governance.
- It integrates open technologies like Apache Iceberg and supports engines such as Spark, Kafka, and Trino.
- An agent-based copilot can translate natural language instructions into data operations and infrastructure tasks.
- Cloudera especially targets private, sovereign AI and agent-based applications with this platform.
The launch arrives as many organizations attempt to move past initial AI experimentation. Building a proof of concept using an API and a limited document set is relatively straightforward. Deploying that same system into production—where data is spread across multiple clouds, internal systems, and legacy platforms—is much more complex.
Cloudera argues that this fragmentation is hindering enterprise projects. The company cites its own research indicating that 73% of surveyed IT leaders believe infrastructure performance limitations have impacted operational initiatives. This figure is based on Cloudera’s studies and should be viewed within the context of the launch’s business environment.
Enterprise AI is beginning to confront the data reality
Anywhere Cloud is built on a simple idea: in many large organizations, the data needed for AI is not located in a single place and cannot be freely moved.
A company might keep some data in a public cloud, other data within its own data center, and certain datasets under specific residency, privacy, or regulatory requirements.
The architecture announced by Cloudera aims to overlay a common layer on these environments. Organizations can independently deploy data and AI services and choose where to run each workload based on technical, economic, or regulatory criteria.
This is especially relevant for the so-called sovereign AI, where the choice of model, execution location, data residency, and infrastructure control all matter.
Cloudera guarantees that certain applications can operate directly on existing repositories without moving or copying data. The company presents this capability as an alternative to large data migrations traditionally involved in modernization projects.
It’s not that any application can run anywhere, unchanged. Feasibility depends on the engine, architecture, computational needs, and data sources. Anywhere Cloud provides the layer that Cloudera aims to simplify for distributing and managing these workloads.
Apache Iceberg, Spark, Kafka, and Trino keep the open component
Another notable aspect is the choice of technologies.
Cloudera Anywhere Cloud uses Apache Iceberg, an open table format for large analytical datasets, with interoperability via Polaris Catalog and unified APIs.
Organizations can also deploy Cloudera engines like Apache Spark, Apache Kafka, or Trino, along with third-party or open-source solutions.
This choice has important implications. A key concern with new enterprise AI platforms is creating a second layer of technology dependence on top of existing cloud and data dependencies.
Cloudera states that its open standards-based architecture aims to minimize this risk and prevent the use of its services from locking data into proprietary formats. Actual interoperability will depend on the specific configurations, components, and services adopted by each organization.
The platform also introduces a self-service model via templates. Teams can deploy certain data, analytics, and AI services without manually configuring the entire underlying infrastructure, while corporate governance policies are enforced on these deployments.
This approach mirrors the abstraction commonly used in public cloud environments but now applied to on-premises setups that may remain physically within the enterprise.
AI agents also start to manage infrastructure
A major innovation linking Anywhere Cloud to the rise of agentic AI is its copilot feature.
Cloudera envisions users describing specific operations in natural language, with the system translating these into workflows on data or infrastructure management tasks.
This represents an interesting evolution of enterprise copilots. The first generation mainly focused on answering questions, code generation, or information retrieval via augmented retrieval methods (RAG). Current agents aim to additionally execute actions using external tools and systems.
This expands their utility but also raises permission, observability, and governance requirements. An agent that can query information poses a different risk level than one authorized to modify infrastructure or execute operations on corporate systems.
Cloudera assures that Anywhere Cloud incorporates a centralized governance model based on zero trust, data traceability, and automatic policy enforcement. These controls become more critical as agents gain access to more enterprise tools.
The company is collaborating with organizations like ADMIRAL Technologies, ExxonMobil, IQVIA, IXEN.ai, and Mastercard in developing Anywhere Cloud. This does not necessarily mean full commercial deployments yet—Cloudera labels these as Customer Design Partners, meaning organizations involved in shaping the platform’s design.
The launch also reflects a broader market shift. For years, much of enterprise modernization revolved around migrating to public cloud. Now, AI is fostering a more distributed architecture.
Computing costs, digital sovereignty, privacy concerns, existing data center investments, and the large volume of accumulated data make bringing AI to the data often more practical than transporting all data to AI systems in certain scenarios.
Anywhere Cloud is Cloudera’s strategy to serve as that intermediary layer.
Frequently Asked Questions
What is Cloudera Anywhere Cloud?
It’s a data and AI platform designed to deploy and manage services across diverse environments, including public cloud, sovereign infrastructure, and private data centers.
Is it necessary to move data to the cloud to use it?
Cloudera suggests that certain workloads can be executed where the data resides, avoiding transfer or copying between infrastructures. The specific capabilities depend on each architecture and application.
What technologies does Cloudera Anywhere Cloud support?
The platform emphasizes open standards and technologies like Apache Iceberg and supports engines such as Spark, Kafka, and Trino, along with third-party solutions.
What does agentic AI contribute to the platform?
Cloudera introduces a copilot capable of interpreting natural language requests to automate specific data workflows and infrastructure tasks.
via: cloudera
