Federal Project Grant Award Summary

Cornell University's Office of Sponsored Programs received a $200,000 Project Grant from the National Science Foundation (NSF) Division of Computing and Communication Foundations under the Computer and Information Science and Engineering program (CFDA 47.070), awarded July 15, 2026, with completion targeted for June 30, 2030. This collaborative research initiative focuses on improving graph-based vector database algorithms to enhance performance, reliability, and computational efficiency. Vector databases are critical infrastructure for storing and enabling efficient search and filtering operations over vector embeddings—numerical representations generated by machine learning (ML) systems from documents, images, videos, and other media. The project aims to develop faster, more resource-efficient algorithms that reduce the computational costs associated with deploying vector database systems.

The deliverables from this research directly address the growing gap between advances in ML/artificial intelligence (AI) and the engineering capability to efficiently leverage vector embeddings at scale. By optimizing core vector database algorithms, the project will enable broader adoption of these systems among resource-constrained users, including small businesses, scientific researchers, and other AI practitioners who lack substantial cloud computing or hardware budgets. Simultaneously, the research will guide the design of next-generation vector database systems capable of managing large-scale datasets, supporting diverse applications across business search, scientific data management, and generative AI systems. The work is being conducted at Cornell University's Ithaca, New York campus.