Does TiDB plan to venture into vectorized databases?

This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: Tidb有打算进军向量化数据库吗?

| username: 将臣17

Does TiDB plan to venture into vectorized databases? For example, text embedding search or image embedding search functionalities?

| username: Billmay表妹 | Original post link

Focus on perfecting the distributed database first!

| username: Timber | Original post link

Is it feasible to use TiKV for storage, with an additional layer of business logic and a retrieval engine on top?

| username: redgame | Original post link

Watch the trend…

| username: Kongdom | Original post link

:+1: :+1: :+1:

| username: 春风十里 | Original post link

What is a quantized database?

| username: srstack | Original post link

Got it!

| username: dcswinner | Original post link

It’s not quantization, it’s vectorization.

| username: 将臣17 | Original post link

Mainly saw these features in Clickhouse, haha, now that the AI era has arrived, we can’t fall behind.

| username: 将臣17 | Original post link

A vector database, simply put, is a database that focuses on distances, such as Euclidean distance, cosine similarity distance, etc. Relational databases are for human use, while vector databases are for AI use.

| username: 将臣17 | Original post link

Like, keep it up!

| username: liuis | Original post link

Similar to the domestic Milvus, of course, currently, ES and Redis also support vector search… In the AI field, it is mainly used to store embedding data.

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.