High-dimensional vector searching method of target database and related equipment

Abstract

The embodiment of the invention provides a high-dimensional vector search method for a target database, communication between a high-dimensional auxiliary process integrated and packaged with a high-dimensional vector search library and the target database is realized through a shared memory, when high-dimensional vector search needs to be performed, search is performed through the high-dimensional auxiliary process, and the search efficiency is improved. Due to integration and packaging in the high-dimensional auxiliary process, the high-dimensional vector searching capability of the high-dimensional vector searching library can be fully utilized, the high-dimensional vector searching speed in the target database can be improved, low-dimensional vector searching or high-dimensional vector searching can be flexibly carried out, and the high-dimensional vector searching efficiency can be improved by expanding the high-dimensional vector searching function of the target database. Therefore, the original database query function and the high-dimensional vector search function of the target database are combined, the vector search efficiency is improved, and the development cost is reduced.

Description

technical field [0001] The invention relates to the field of data search, in particular to a high-dimensional vector search method of a target database and related equipment. Background technique [0002] In the Internet era, the scale of unstructured data has grown rapidly, such as video, images, voice, text, etc. With the rise of deep learning methods in various fields, most unstructured data can be converted into high-dimensional vectors for structural These vector data can be effectively used through the vector search method. However, in practical applications, these data are stored in databases. Due to the large time span of data resources in databases, most databases are developed based on low-dimensional vectors, so they only support low-dimensional vector searches, such as PostgreSQL's imgsmlr extension and cube extension store high-dimensional structured data in the PostgreSQL database, and cannot directly use existing vector search methods for high-dimensional vec...

AI Technical Summary

Problems solved by technology

However, in practical applications, these data are stored in databases. Due to the large time span of data resources in databases, most databases are developed based on low-dimensional vectors, so they only support low-dimensional vector searches, such as PostgreSQL's imgsmlr extension and cube extension store high-dimensional structured data in the PostgreSQL database, and cannot directly use existing vector search methods for high-dimensional vector search

Images