VectorDB CLI + VectorDB Lite 本地高效调试向量数据库

更新时间：2025-04-08

使用Docker Compose启动VectorDB服务

启动方式 1

Docker Compose配置docker-compose.yml

                Shell
                
            

                services:
  vdb-service:
    image: mochow/vdb:2.1.5854  # 镜像名称和版本
    container_name: baidu-vdb-container  # 容器名称
    ports:
      - "5287:5287"  # 映射容器内的 5287 端口到主机
    environment:
      - port=5287  # 配置容器内使用的端口
    volumes:
      - ./data:/mnt/data  # 将本地 ./data 目录挂载到容器中的 /mnt/data
      - ./log:/mnt/log  # 将本地 ./log 目录挂载到容器中的 /mnt/log
    restart: always  # 自动重启策略，确保容器在意外退出时重启
            

启动本地VectorDB服务

Shell

1docker compose up -d

启动方式 2

国内用户如果无法访问 docker 服务，可以直接下载，然后通过命令安装

Shell

1curl http://public-vdb.bj.bcebos.com/vdb-standalone-2.1.tar.gz -o vdb-standalone-2.1.tar.gz

解压后启动本地 VectorDB 服务

Shell

1sh vdb_service.sh start

启动后本地会有两个目录，data保存数据和log保存日志

                Shell
                
            

                ll
total 12
drwxr-xr-x 5 1000 1000 4096 Jan 10 10:57 data
-rw-r--r-- 1 root root  536 Jan 10 10:55 docker-compose.yml
drwxr-xr-x 2 1000 1000 4096 Jan 10 10:57 log
            

查看服务正常启动，服务端口为5287

                Shell
                
                docker ps | grep vectordb
709db1519963   mochow/vdb:2.1.5854                "/root/entrypoint.sh"    About a minute ago   Up About a minute         0.0.0.0:5287->5287/tcp, :::5287->5287/tcp                                              baidu-vdb-container

停止本地VectorDB服务

Shell

1docker compose down

使用VectorDB CLI进行建表

使用VectorDB CLI链接本地的VectorDB，root默认密码为mochow。

                Shell
                
            

                ./vectordb-cli -u root -e http://127.0.0.1:5287
api key for 'root' to 'http://127.0.0.1:5287': 
2024/09/14 11:06:53 Successfully initialized mochow client.


        __     __                _                    ____    ____  
        \ \   / /   ___    ___  | |_    ___    _ __  |  _ \  | __ ) 
         \ \ / /   / _ \  / __| | __|  / _ \  | '__| | | | | |  _ \ 
          \ V /   |  __/ | (__  | |_  | (_) | | |    | |_| | | |_) |
           \_/     \___|  \___|  \__|  \___/  |_|    |____/  |____/


        type '-h' or 'help' to see usage
vectordb-cli > 
vectordb-cli > 
            

建库

                Shell
                
            

                vectordb-cli > create database -d testdb
create database 'testdb' success
vectordb-cli > list databases
+--------------+
| databases    |
+==============+
| testdb       |
+--------------+
            

建表

                Shell
                
            

                vectordb-cli > create table

input create table args, end with ';'
... {
...      "database": "testdb",
...      "table": "book",
...      "description": "basic test",
...      "replication": 1,
...      "partition": {
...          "partitionType": "HASH",
...          "partitionNum": 3
...      },
...      "enableDynamicField": false,
...      "schema": {
...          "fields": [
...              {
...                  "fieldName": "id",
...                  "fieldType": "STRING",
...                  "primaryKey": true,
...                  "partitionKey": true,
...                  "autoIncrement": false,
...                  "notNull": true
...              },
...              {
...                  "fieldName": "bookName",
...                  "fieldType": "STRING",
...                  "notNull": true
...              },
...              {
...                  "fieldName": "author",
...                  "fieldType": "STRING"
...              },
...              {
...                  "fieldName": "page",
...                  "fieldType": "UINT32"
...              },
...              {
...                  "fieldName": "vector",
...                  "fieldType": "FLOAT_VECTOR",
...                  "notNull": true,
...                  "dimension": 3
...              }
...          ],
...          "indexes": [
...              {
...                  "indexName": "book_name_idx",
...                  "field": "bookName",
...                  "indexType": "SECONDARY"
...              },
...              {
...                  "indexName": "vector_idx",
...                  "field": "vector",
...                  "indexType": "HNSW",
...                  "metricType": "L2",
...                  "params": {
...                      "M": 32,
...                      "efConstruction": 32
...                  }
...              }
...          ]
...      }
...  };
create table 'book' success

vectordb-cli > list tables --db testdb
+-----------+
| tables    |
+===========+
| book      |
+-----------+
            

操作指南

AI Search 高效构建 RAG 服务

百度智能云

向量数据库 VectorDB

向量数据库 VectorDB

VectorDB CLI + VectorDB Lite 本地高效调试向量数据库

使用Docker Compose启动VectorDB服务

启动方式 1

启动方式 2

使用VectorDB CLI进行建表