en.Wedoany.com Reported - Everpure (formerly Pure Storage) unveiled a data intelligence platform at its annual Pure Accelerate conference, introducing a "data fabric" solution. Enhanced based on the flagship platform of 1Touch, which was recently acquired, this solution aims to help enterprises move away from application-centric data silos toward a governed, AI-ready data foundation.

The solution crawls SaaS applications, databases, object or file storage, and mainframes to build a cross-environment context layer containing an index and semantic graph, recording where data resides, its content, and the relationships between different records. It also applies governance, security, and residency policies, feeding so-called "AI-ready" views containing only relevant data into downstream systems and tools, ensuring enterprise customers do not use raw or context-free unprocessed data directly for their AI models or agents.

Prakash Darji, General Manager of Everpure's Digital Experience business, stated at a press conference that the solution aims to help people leverage and adopt AI in a secure, governed, and cost-effective manner, with scalability. Darji directly compared the direction of data intelligence to API management, calling it the API gateway of the data era. He emphasized that data intelligence acts more like a neutral orchestration and control layer, capable of "liberating" information from various applications to form a shared and governed system of record.

The solution's automated governance capabilities can scan systems to identify potential sensitive information and track related lineage. Model Context Protocol support is another differentiating advantage of the platform, which Darji described as a natural language means to better understand where data resides and ensure security rules govern which agents or applications can access data, ultimately mapping raw data to business definitions through a semantic knowledge graph.

When comparing the market landscape, Darji publicly expressed skepticism about the current AI data landscape, noting that vendors like Databricks, SAP, and Salesforce are all making the same claim of bringing data into their specific ecosystems. He believes data will not be controlled on a single platform and does not subscribe to arguments in self-serving vendor camps. Everpure positions data intelligence as a heterogeneous solution that can run where data resides. Although the launch is compared to Databricks Unity Catalog or Snowflake Horizon, Darji pointed out that Databricks itself is also seen as a source for the platform, and the two have collaborated on Apache Iceberg open table sharing.

The data intelligence solution heavily relies on the acquisition of 1Touch. Ashish Gupta, former CEO of 1Touch and now General Manager of Data Management at Everpure, stated that 1Touch brought the "context layer" that Pure lacked, which he considers the core competitive moat. Gupta said the solution does more data discovery than other companies in the market, contrasting with DSPM tools that only cover cloud environments. What 1Touch adds is inference-based classification, building an understanding of how data is actually used in business processes. This depth makes the solution relevant not only for security and compliance use cases but also provides behavioral telemetry to drive AI applications.

Everpure executives emphasized that while the company is newly focused on data intelligence and governance, it will not abandon its traditional storage foundation business. Rob Lee, Chief Technology and Growth Officer at Everpure, compared this shift to "storage, and." He explained that the rebranding, company name change, and new direction focusing on deeper data intelligence, data governance, classification, and understanding were the result of collaborative work over the past 18 months, coinciding with the timing of the 1Touch acquisition. Darji further speculated that semantic knowledge graphs could ultimately solve a more fundamental AI infrastructure problem, noting that due to rising costs and accuracy issues caused by expanding context windows, ontologies and knowledge graphs could effectively serve as persistent memory for Agent AI if they keep parsed information closer to the underlying data, solving the context window problem just as flash memory once solved memory limitations.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com