forked from postman/cleanproject
63 lines
1.4 KiB
Plaintext
63 lines
1.4 KiB
Plaintext
|
# mysql sysbench
|
|||
|
|
mysql sysbench是一款压测数据生成工具
|
|||
|
|
|
|||
|
|
## python3 生成压测数据说明
|
|||
|
|
### quick_insert.py
|
|||
|
|
|
|||
|
|
使用程序方式批量插入数据的脚本，需要引入依赖包 pymysql
|
|||
|
|
|
|||
|
|
### quick_insert_threads.py
|
|||
|
|
|
|||
|
|
多线程批量插入数据的脚本，需要引入依赖包 pymysql
|
|||
|
|
|
|||
|
|
参数配置说明：
|
|||
|
|
|
|||
|
|
```
|
|||
|
|
写入总数为：threads * times * sql_batch
|
|||
|
|
offset : 偏移量，默认是0。如果数据库有数据，可以设一个起始点（偏移量）
|
|||
|
|
threads : 线程数
|
|||
|
|
times : 每个线程执行commit的次数
|
|||
|
|
sql_batch : 每次commit多少数据
|
|||
|
|
例：要造一亿数据
|
|||
|
|
threads = 5
|
|||
|
|
times = 2000
|
|||
|
|
sql_batch = 10000
|
|||
|
|
```
|
|||
|
|
|
|||
|
|
### quick_sql_insert.py
|
|||
|
|
|
|||
|
|
生成 insert 批量插入脚本，与上面的方式类似，只不过生成的文件需要手动执行。
|
|||
|
|
|
|||
|
|
### quick_txt_load.py
|
|||
|
|
|
|||
|
|
生成逗号分隔的 txt 文件，之后需要使用 `load data infile` 命令导入数据。
|
|||
|
|
|
|||
|
|
|
|||
|
|
## 使用方法
|
|||
|
|
### 配置数据源
|
|||
|
|
|
|||
|
|
```yaml
|
|||
|
|
host = "127.0.0.1"
|
|||
|
|
port = 3306
|
|||
|
|
username = "root"
|
|||
|
|
password = "123"
|
|||
|
|
charset = "utf-8"
|
|||
|
|
db = "test"
|
|||
|
|
```
|
|||
|
|
### 启动单线程版
|
|||
|
|
|
|||
|
|
```shell
|
|||
|
|
python3 quick_insert.py &
|
|||
|
|
```
|
|||
|
|
|
|||
|
|
### 启动多线程版
|
|||
|
|
|
|||
|
|
```shell
|
|||
|
|
python3 quick_insert_threads.py &
|
|||
|
|
```
|
|||
|
|
|
|||
|
|
# 说明
|
|||
|
|
***测试可以直接使用sql/tab.sql建表，也可以改成自己的表结构***
|
|||
|
|
|
|||
|
|
- [生成mysql亿级压测数据程序](https://blog.csdn.net/evane1890/article/details/112911724)
|