[Feature Request]: Integrating MatrixOne as a vector store method

[wknet123]

Feature Description

• Support vector data store by automatic creating or re-creating a MatrixOne table with specified embedding dimension size which generated by common embedding model such as m3e, bge etc.
• Support vector retrieval by integrating llamaindex VectorStoreIndex with MoVectoRs instance to its vector_store paramater.
• Support vector data store deletion from table.
• Support deployment as cloud service or standalone mode.

Reason

MatrixOne, as a relational database with vector capabilities, provides powerful data management capabilities.
LlamaIndex does not support MatrixOne as vector store integration currently.
Matrixone supports vector retrieval using the following distance measure functions:

• Cosine similarity function cosine_similarity
• Cosine distance function cosine_distance
• L2 distance function l2_distance

Value of Feature

• MatrixOne is a hyper-converged cloud & edge native distributed database with a structure that separates storage, computation, and transactions to form a consolidated HSTAP data engine. This engine enables a single database system to accommodate diverse business loads such as OLTP, OLAP, and stream computing. It also supports deployment and utilization across public, private, and edge clouds, ensuring compatibility with diverse infrastructures.
• Matrixone currently supports vector retrieval by using distance measure functions.
• Make llamaindex integrating with MatrixOne as a vector store method would provide developer a new option to support AIGC application development, not only for vector data but also for business data that MatrixOne touts significant features, including real-time HTAP, multi-tenancy, stream computation, extreme scalability, cost-effectiveness, enterprise-grade availability, and extensive MySQL compatibility.