ModaHub Community: Will vector databases face selection problems like traditional databases? Can you introduce what kind of enterprise is suitable for what type of vector database? Is it a stand-alone version, a plug-in version, or an enterprise-level version?
Li Li's point of view is very clear. He believes that when selecting a model, the first thing to consider is the demand. If the business volume is not particularly large and there are cost considerations, you can use the method of combining two ready-made technologies, which is the fastest and most trouble-free method. The advantage of this method is that it is simple and easy to implement. It does not require a lot of time and resources to develop new technologies. It only needs to properly transform and integrate existing technologies to meet current needs.
Specifically, this combination method can be performed according to the user's usage habits. For example, if a company particularly likes to use MySQL, the capability of vector engine can be added to MySQL, so that MySQL has the capability of vector search. Likewise, if a company particularly likes to use PostgreSQL, it is also possible to add vector engine capabilities to PostgreSQL. In addition, other types of databases such as Redis or ElasticSearch can also be considered, and the capabilities of vector engines can also be added to it.
For small-scale companies, this combination is very suitable. Because small-scale companies often do not have too many resources to invest in new technology research and development, and their business level is not enough to support them to develop a brand new vector database. Therefore, combining existing technologies can well meet their needs.
</