Cloudera introduces Anywhere Cloud, a platform that enables companies to build and run data and AI applications in public clouds, sovereign infrastructure, and their own data centers. Everything runs through a single control plane, without moving data.
Cloudera’s new platform aims to resolve a problem many organizations recognize: AI projects that get stuck between the experimental and production phases. According to Cloudera, this stems from fragmented infrastructure, slow upgrade cycles, and increasingly stringent data sovereignty requirements. The company cites research showing that 73 percent of IT leaders believe that performance limitations in the infrastructure are hindering their projects.
Anywhere Cloud can run independently or be deployed alongside existing data lakes. Its modular architecture lets organizations roll out and manage individual data and AI services through a single control plane. Cloudera promises this will help avoid migrations and vendor lock-in.
Modules from a marketplace
Customers select services via self-service blueprints in a marketplace. These can include engines such as Spark, Kafka, or Trino, as well as open-source or partner components. Interoperability is enabled through Apache Iceberg, the Polaris catalog, and uniform APIs.
The platform also includes an agent-based copilot. It translates plain-language requests into automated data workflows or infrastructure management. Governance remains centrally managed according to a zero-trust model, with continuous lineage across distributed data environments.
“Enterprise AI has outgrown the public cloud-only model,” says Leo Brunnick, Chief Product Officer at Cloudera. According to him, organizations don’t have to choose between innovation and control, and Anywhere Cloud brings cloud speed wherever business data resides.