CTERA presents at AI Infrastructure Field Day 3

CTERA Intelligent Data Management from Edge to Core

Watch on YouTube
Watch on Vimeo

Discover how CTERA addresses the complexities of hybrid cloud storage by enhancing operational efficiency and security, advocating a unified platform that extends from edge to cloud to manage increasing data demands. Practical use cases across various industries demonstrate how CTERA leverages AI-powered tools, automated workflows, and an intelligent data platform to improve data insights, security, and governance on a wide scale.

CTERA positions itself as a leader in distributed data management, providing a file system fabric that connects edge locations, core data centers, and cloud environments. Their platform allows for secure data access, protection, and sharing, forming the foundation for future AI applications. CTERA’s DNA comes from the security space and works with highly regulated industries such as healthcare, financial services, and government agencies. These organizations share common challenges of data cybersecurity consciousness, highly distributed data, and the need for information management.

CTERA addresses the increasing complexity of hybrid cloud storage by offering a platform that connects data centers, cloud providers, and edge offices. The company uses object storage as its backbone, providing reliability, durability, capacity, and cost-effectiveness. To overcome the limitations of centralized object storage, CTERA adds a complementary layer that caches data, provides performance, and enables multi-protocol access across various locations. The company modernizes unstructured data across operational efficiency, cyber storage and proactive data protection, productivity through multiprotocol access, and generative AI to improve productivity and security.

CTERA’s architecture consists of edge appliances that present file shares, optimize performance through caching, and provide real-time data protection to a centralized object store. This model enables near zero-minute disaster recovery and facilitates data sharing across geographically dispersed locations. CTERA simplifies data migration from traditional NAS platforms using built-in tools, allowing customers to move large datasets to CTERA-managed object storage, which can reside on-premise or in the cloud. The company excels in industries such as healthcare, public sector, retail, and manufacturing, where security, data distribution, and the need for modernization are critical.

Personnel: Saimon Michelson

CTERA Enterprise Intelligence: Unify & Activate Your Private Data for Faster & Smarter Outcomes

Watch on YouTube
Watch on Vimeo

This session offers insight into the seamless integration of AI within the CTERA Intelligent Data Platform. Embedded AI and analytic Enterprise Data Services are explored, along with the underlying data fabric that facilitates secure, global data connectivity and ensures high-performance access. Additionally, the session demonstrates how agentic AI advances productivity, enabling virtual employees to interact efficiently with private data resources while maintaining robust security and operational effectiveness.

CTERA’s approach to enterprise intelligence is structured around three key pillars. The first pillar focuses on embedding AI and analytics within data management for enhanced security and data quality. This includes real-time inspection of I/O for anomaly detection and integration with security solutions like Purview. The second pillar involves providing a data foundation and fabric for AI training and inferencing, offering a global namespace for data aggregation and access, along with real-time metadata notifications and direct object access via a direct protocol that bypasses the CTERA portal.

The third pillar introduces agentic AI, enabling virtual employees to interact with private data resources efficiently. This pillar provides a semantic layer, allowing connections to various data sources (including non-CTERA systems), enrichment of metadata, and normalization of permissions across different data sources. This facilitates semantic searches and retrieval of relevant documents, empowering users with productivity-enhancing tools within a secure environment and answering questions with a chat-based interface.

Personnel: Saimon Michelson

Unlocking Enterprise Intelligence with CTERA MCP

Watch on YouTube
Watch on Vimeo

In this session, learn how the Model Context Protocol (MCP) tackles the challenges of utilizing unstructured data by providing seamless, permission-aware integration for AI models and data sources, eliminating the need for intricate custom connectors. Discover how this ‘USB for AI’ enables enterprise-wide data interaction and management, offering a reliable and future-proof architecture.

CTERA addresses the problem of connecting enterprise data sources to Generative AI models, which traditionally required custom connectors for each application, resulting in exponential complexity and fragility. The MCP protocol offers a solution by providing a seamless, guaranteed integration between any Gen AI model and tool that supports MCP, while also being permission and identity aware. It ensures contextual information about the user, their permissions, and authentication is readily available. CTERA has embraced MCP as a core part of its strategy, implementing it across its products.

CTERA’s implementation of MCP is structured in two main layers. The first layer resides within the global file system product, where files are stored, enabling Gen AI agents to access and utilize data while respecting user permissions. The second layer focuses on data intelligence, providing a semantic layer over the data that exposes textual information and metadata through MCP. The MCP server is implemented within the enterprise application, while the MCP client is the AI tool. This architecture is not specific to any LLM and supports OAuth2 authentication, allowing for secure and permissioned access to data.

A demonstration highlighted how CTERA’s MCP server could be easily enabled via the user interface, showcasing its integration with Claude. The demonstration showed how a user could instruct Claude to interact with the global file system, list files, read them, summarize them, and write the summary back, all without writing any code. This example illustrated how MCP enables end-to-end applications that democratize access to data and allow users to simplify repetitive tasks, thereby increasing efficiency and job satisfaction.

Personnel: Aron Brand

From Storage to Enterprise Intelligence, Unlock AI Value from Private Unstructured Data with CTERA

Watch on YouTube
Watch on Vimeo

Discover the obstacles that hinder AI adoption. What matters most? Data quality or quantity? Understand the strategy CTERA uses for curating data to create trustworthy, AI-ready datasets that overcome silos and security challenges, translating raw data into meaningful and actionable insights.

CTERA’s presentation at AI Infrastructure Field Day 3 focused on the transition from traditional storage solutions to “enterprise intelligence,” highlighting the potential of AI to unlock value from unstructured data. While enterprise GenAI represents a massive market opportunity, with projections reaching $401 billion annually by 2028, the speaker, Aron Brand, emphasized that current adoption is hindered by the poor quality of data being fed into AI models. Brand argued that simply pointing AI tools at existing data leads to “convincing nonsense,” as organizations often lack understanding of their own data, resulting in inaccurate and potentially harmful outputs. He identified three main “quality killers”: messy data, data silos, and compliance/security concerns.

To overcome these obstacles, CTERA proposes a strategy centered on data curation, involving several key steps. These include collecting data from various storage silos, unifying data formats, enriching metadata, filtering data based on rules and policies, and finally, vectorizing and indexing the data. CTERA aims to provide a platform that enables users to create high-quality datasets, enforce permissions and guardrails, and deliver precise context to AI tools. The platform is powered by an MCP server for orchestration and an MCP client for invoking external tools, facilitating an open and extensible system.

CTERA’s vision extends to “virtual employees” or subject matter experts created by users to automate tasks and improve efficiency. The system respects existing access controls and provides verifiable answers grounded in source data. The presented examples demonstrated the potential of the platform in various use cases, including legal research, news analysis, and medical diagnostics. The presentation emphasized that the goal is not to replace human workers but to augment their capabilities with AI-powered assistants that can access and analyze sensitive data in a secure and compliant manner.

Personnel: Aron Brand