Data Engineering · 3 min read
Hitachi Vantara Lets Object Storage Do the Analytics, Up to 5x Faster
VSP One Object now queries Iceberg, Parquet, JSON and CSV in place, with up to 5x faster analytics, 80% fewer CPUs and a new four-node entry configuration for AI teams.
Key takeaways
- VSP One Object queries Iceberg, Parquet, JSON and CSV in place, with analytical queries up to 5x faster and 80% fewer CPUs
- Small-object performance improves up to 5.5x for AI, analytics and S3 workloads
- A new four-node entry configuration lowers the deployment starting point
- iQ Studio adds NFS and filesystem knowledge sources, a Model Gateway, MCP connectivity and ServiceNow support
- McKinsey: 44% of organisations scale AI enterprise-wide; IDC: 89% report data-quality problems

Hitachi Vantara claims analytical queries run up to five times faster on its updated VSP One Object platform, while burning 80% fewer CPUs, by letting users query data where it already sits. Announced 18 November 2026, the expansion targets the least glamorous bottleneck in enterprise AI: moving copies of data between storage, analytics and model systems until nobody trusts any of them.
What actually changed in VSP One Object?
The headline feature is in-place querying. VSP One Object now supports direct queries against Iceberg, Parquet, JSON and CSV formats held in object storage, skipping the usual ETL shuffle into a separate analytics engine. Hitachi pitches this as fewer duplicate datasets, shorter data pipelines and less supporting infrastructure. In other words, the storage array wants to be the lakehouse.
Two more numbers deserve attention:
- Small-object performance improves up to 5.5x for AI, analytics and S3 workloads, which matters for retrieval-heavy pipelines rather than batch scans.
- A four-node entry configuration lowers the starting point for AI and analytics deployments, aimed at teams that couldn't previously justify the footprint.
The release also adds expanded multisite replication and regional capabilities for availability and data protection across locations. Sensible, unglamorous, needed.
What does iQ Studio add?
Hitachi iQ Studio, the software layer for building, deploying, managing and governing AI applications and agents, gets broader access to structured and unstructured enterprise data for retrieval-augmented generation. New filesystem and NFS knowledge sources mean organisations can point AI systems at existing repositories without first moving or restructuring the data. Anyone who has watched a RAG project stall on "we just need to copy it first" will recognise the appeal.
The rest of the list reads like a checklist for governed AI deployments: a Model Gateway for local and external models, Model Context Protocol connectivity, ServiceNow support, role-based access, observability and lifecycle management tools. The governance features aren't decoration. IDC research cited by Hitachi puts 89% of organisations as reporting some level of data-quality problem, and access control over models, tools and data is attracting growing scrutiny.
Is the "experimentation to production" framing credible?
Hitachi leans on McKinsey figures showing nearly nine in 10 organisations regularly use AI in at least one business function, while only 44% say AI is scaling across the enterprise. That gap is the sales pitch, and it's a fair one. Chief Product Officer Octavian Tanase framed the release as tying data management to AI deployment directly.
"AI strategy is now inseparable from data strategy," Tanase said. "As AI becomes part of more business-critical workloads, the infrastructure underneath it has to make it easier to put enterprise data to work."
He added: "With VSP One, we're giving customers a practical way to build the data foundation for AI, bring analytics closer to their data and scale with confidence as demand grows. That creates a stronger path from experimentation to production without sacrificing the resilience and control enterprises depend on."
What should data teams do with this?
The strategic read: storage vendors are done being passive custodians. Hitachi is packaging object storage, direct analytics and governed model access in one stack under a common control plane, where VSP One Object handles storage and Hitachi iQ connects AI applications, agents, models, tools and workflows to enterprise data.
The filesystem, NFS and external model connectivity suggests the target buyer runs a fragmented estate spanning older and newer repositories. If your organisation matches that description, the four-node entry point makes a proof-of-concept cheap enough to test the 5x query claim against your own workloads. Verify before believing. Vendor benchmarks are run under laboratory conditions, and your Iceberg tables won't be.
For teams already committed to a separate lakehouse architecture, this is watch-and-wait. For teams drowning in pipeline sprawl, an in-place query layer with 80% fewer CPUs is at least worth a benchmark. The industry-wide shift from storing data to using it in place isn't new. Hitachi's version is merely the latest, and better-numbered, entry.
Google News: Data Governance (Source)
More from Grace Kim
Show full bio
Market editor covering companies, deals and strategy in Analytics Digest is a professional daily at Analytics Digest.
18 articles