Object Storage Moves AI Queries Closer to Enterprise Data

Hitachi Vantara announced on November 11, 2026, that VSP One Object can query supported open-format data directly on the storage platform. The update gives organizations a four-node starting configuration, creating a smaller entry point for AI-ready object storage while simplifying the first stage of deployment.

The announcement connects three parts of the enterprise AI stack: object storage, lakehouse analytics, and the software used to build and govern AI applications and agents. VSP One provides the data foundation for Hitachi iQ, which combines validated AI infrastructure and software to connect applications and agents with enterprise data, models, tools, and workflows.

The new capabilities also add governance features to Hitachi iQ Studio, the enterprise software for building, deploying, managing, and governing AI applications and agents. Together, the changes focus on how organizations store data, query it for analysis, and use it to support AI systems.

Queries Without Moving Data

VSP One Object supports direct queries against Iceberg, Parquet, JSON, and CSV data. That means supported files can be analyzed on the object-storage platform instead of being copied into separate analytics systems first.

Running queries in place can reduce duplicate datasets, data pipelines, and the infrastructure needed to maintain them. It also gives organizations a closer link between the data being analyzed and the storage platform holding that data, which matters when AI applications and analytical tools depend on current enterprise information.

Hitachi Vantara reports up to 5x faster analytical query performance with 80% fewer CPUs from the new capabilities. The company also reports that continued small-object performance improvements support up to 5.5x faster data access for AI, analytics, and S3 workloads.

Those figures address two different parts of the workload. Faster analytical queries focus on how quickly systems can examine data, while improved small-object access supports the many small files and objects that AI, analytics, and S3 workloads can use.

Why In-Place Analytics Matters for AI

Moving data into separate systems can create governance problems as well as extra infrastructure. Copies can drift, so an explanation generated from an older or differently filtered extract may not match the data behind a prediction.

Analyzing the same data in place keeps the explanation and prediction aligned. It also preserves a clearer lineage trail through the tables, columns, and transformations that fed a model. That trail helps show how data moved into an AI result without requiring another copy of the records.

Hitachi Vantara presents this approach as relevant to energy and medical workloads. In grid analysis, one example joins sensor time series with asset, outage, and billing data. In medical work, another example queries imaging metadata and structured reports without duplicating sensitive records into separate analytics environments.

For AI and analytics workloads spread across sites, data access also involves what happens during a disruption. Expanded multisite replication and regional capabilities are designed to keep data available, protected, and accessible when something goes wrong.

That requirement differs from query speed. Faster queries help systems work through data sooner, while replication and regional capabilities focus on continued access to the underlying data during disruptions.

More Governance for Enterprise AI Applications

Hitachi iQ Studio receives new filesystem and NFS knowledge sources, expanding access to structured and unstructured enterprise data for retrieval-augmented generation, or RAG. RAG supplies enterprise context for AI responses and agents without requiring organizations to move or restructure all of that data first.

This gives applications and agents a way to use information across existing enterprise data sources. The change connects the data foundation in VSP One with the software layer that manages AI applications and agents, while keeping the focus on access, context, and governance.

A new Model Gateway governs access to local and external models. That adds a control point for how AI applications and agents reach the models they use, whether those models run locally or come from external sources.

The four-node configuration rounds out the update by lowering the starting point for AI-ready object storage. Simplified deployment is intended to make initial setup easier, with capacity able to scale as demand grows.

Put together, the changes aim to reduce the distance between enterprise data and the AI systems that use it. VSP One Object handles direct queries across supported open formats, while Hitachi iQ Studio adds data access and governance features for the applications and agents built on top of that foundation.

Stay connected via Google News

Follow us for the latest AI News