**London**: Teradata has unveiled its Enterprise Vector Store, designed for efficient vector data management in hybrid cloud environments. The solution supports Trusted AI, integrates with NVIDIA technology, and aims to streamline unstructured data processing for enhanced business intelligence and customer service.
Teradata has introduced an in-database solution called the Enterprise Vector Store, aimed at enhancing vector data management within its hybrid cloud platform. This new offering supports the growing demand for Trusted AI, with plans for future integration of NVIDIA’s NeMo Retriever microservices, part of the NVIDIA AI Enterprise software suite.
The Enterprise Vector Store enables organisations to process billions of vectors and integrate them into existing enterprise systems. Response times are reported to be as fast as tens of milliseconds, making the solution cost-effective for addressing complex business challenges. By providing a single, trusted repository for all data, it builds on Teradata’s current support for retrieval-augmented generation (RAG) technologies, essential for dynamic AI use cases.
Louis Landry, Chief Technology Officer at Teradata, emphasised the significance of vector stores in enabling effective application of generative AI and agentic AI, stating, “Vector stores are at the root of how we bind truth to generative AI models and agentic AI. They are essential to any data management practice, but their impact is limited when they are slow or siloed.” He highlighted that Teradata’s expertise allows the Enterprise Vector Store to deliver the necessary foundation for large organisations seeking to implement these advanced AI systems.
The solution aims to facilitate capabilities that require managing unstructured data—such as text, images, and videos—alongside structured data for comprehensive analysis. Notably, it incorporates features that support the whole lifecycle of vector data management and adheres to data governance practices critical for Trusted AI.
The Enterprise Vector Store is designed with the scalability needed for current business requirements, leveraging Teradata’s hybrid environment to function seamlessly across cloud and on-premises infrastructures. It will also be compatible with leading frameworks like LangChain and RAG, and plans to include temporal vector embedding capabilities to enhance trust and explainability by tracking changes over time.
Additionally, Teradata is set to enhance the Enterprise Vector Store by integrating NVIDIA’s NeMo Retriever, which will improve information retrieval from unstructured sources such as PDFs. Pat Lee, Vice President of Strategic Enterprise Partnerships at NVIDIA, noted, “Data is essential to accurate inference for AI applications. Teradata Enterprise Vector Store, integrated with NVIDIA AI Enterprise and NVIDIA NeMo Retriever, can unlock the institutional knowledge stored in PDFs and other unstructured documents to power intelligent AI agents.”
The application of the Enterprise Vector Store is illustrated in the context of augmented call centres, where AI agents can rapidly deliver tailored customer service. For instance, an insurance company’s multi-agent system could quickly access harmonised data to answer specific customer inquiries with precision, improving service efficiency.
The Teradata Enterprise Vector Store is currently available in private preview, with widespread access anticipated in July. The deployment of this technology indicates a significant move towards maximising value from unstructured data while optimising costs, particularly for enterprises looking to adopt advanced AI solutions in their operations.
Source: Noah Wire Services
Subscribe to Industry Updates
