Quickly update the information in a field in kettle

There is an order table in the database tb_order, Among them is order_id and user_id and user_name Etc .
There are user data tables in the database tb_user, Among them is user_id and user_name Etc .
Now you need to use tb_user.user_name To update tb_order.user_name, The association condition of the two tables is tb_order.user_id = tb_user.user_id.
Usually , When both watches are stationary , You can use one update To solve , as follows .

update tb_order
set user_name = (select user_name from tb_user u where u.user_id = tb_order.user_id);

however , If the data volume of these two tables is large , And both tables are frequently used in production , One update Statement will lock the table , And it takes a long time , Thus, normal business may not be carried out quickly .
here , It should be traversed separately tb_order Every record in the table , And then according to tb_order.user_id Go to tb_user Query the user_name, According to the tb_order.order_id To update tb_order.user_name; This way of traversal alone , Although it can't completely solve the problem caused by locking the table , But it can also be avoided with a high probability . But this configuration mode , Very inefficient .
There is a better idea , Configuration that can be completed by a transformation . Ideas as follows ：
1、 Use table input , from tb_order Data from the table ;
2、 Use the database connection control , Get parameters directly from the results entered in the table , Then execute the query , from tb_user Table get the information you need ;
3、 Use insert to update space , Directly modifying tb_order The fields specified in .

Because these three steps are in a transformation , therefore , In the first 2 Set the number of concurrency in step and step to improve kettle The processing efficiency of .
The configuration is as follows ：