Hitachi Vantara Enhances VSP One Object for AI and Analytics Workloads, Streamlining Data Lakehouse Operations

Hitachi Vantara has announced significant enhancements to its Virtual Storage Platform One (VSP One) Object, focusing on optimizing it for enterprise AI and analytics workloads. Key updates include the ability to directly query open-format data (such as Iceberg, Parquet, JSON, and CSV) on the storage platform itself, and the introduction of a smaller four-node starting configuration for easier deployment and scalability. The company reports these new capabilities can lead to up to 5x faster analytical query performance with 80% fewer CPUs. This development is significant for organizations increasingly relying on data lakes and lakehouses for their AI and analytics initiatives. By allowing direct querying on the object storage layer, Hitachi Vantara is tackling the common challenge of data movement, which often introduces latency, complexity, and additional costs. Data practitioners can now reduce the need to copy or move data to separate analytics systems, streamlining their data pipelines and improving governance by maintaining data lineage. The smaller starting configuration also lowers the barrier to entry for organizations looking to deploy AI-ready object storage, enabling them to start small and scale as their data needs grow. This move aligns with a broader trend in cloud and data management where object storage is evolving from a passive data repository to an active, intelligent data layer. Cloud providers and storage vendors are increasingly integrating compute capabilities directly with storage, often leveraging open table formats like Apache Iceberg, to enable in-place analytics. This paradigm shift aims to bring computation closer to the data, minimizing data egress costs and improving performance for data-intensive workloads like AI model training and inference. Other vendors are also focusing on high-performance object storage and smart storage features that integrate AI for automated metadata annotation and improved data management. In practice, these enhancements mean that data engineers and data scientists can expect a more agile and cost-effective approach to building and managing their data lakehouses. The ability to query data directly on VSP One Object reduces the operational burden of managing complex ETL processes and duplicate datasets. Practitioners should evaluate how these features can be integrated into their existing data architectures, particularly for workloads that benefit from reduced data movement and faster query times. It also highlights the importance of open data formats in enabling greater flexibility and interoperability across different analytics tools and platforms. Organizations should consider the trade-offs between specialized analytics platforms and the growing capabilities of object storage for in-place analytics when designing their data strategies.