Prerequisite: None
Vector databases are essential in a world overflowing with complex data sets. They have come into vogue in the past few years as core infrastructure for generative artificial intelligence (AI). They help users discover complex patterns of similarity, affinity, and connectedness in unstructured and multimodal data sets, such as those containing text, images, video, and audio. This enables generative applications to recombine the data in new ways in response to user-input prompts.
Within a modern data strategy, vector databases can support a wide range of use cases beyond generative AI. They have a long track record powering the advanced analytics at the heart of recommendation engines, similarity search, influence analysis, and other essential enterprise applications.
TDWI senior research director James Kobielus will explore the key issues and criteria that enterprises should consider when evaluating vector databases within their data modernization strategies. He will cover such considerations as:
- What are vector databases?
- What are the principal use cases of vector databases?
- What commercial and open source vector databases are available now?
- Should enterprises adopt a premises-based, cloud-hosted, or SaaS-based vector database?
- What are the cost, performance, scalability, manageability, programmability, and other key factors to consider when evaluating vector databases?
- How do vector databases complement and integrate with other types of databases in a modern enterprise data architecture?
- What skills, tools, and frameworks do data professionals require to work with vector databases?