Load Method

The Data Loader has three different methods of loading data.

You would use whichever one runs fastest for your database type, network speed and other factors. If doing large loads you may wish to run some test loads to determine the method which runs fastest in your environment.

Default Method

The default load method is to use a parameterized Insert statement. Example:

insert into table (col1, col2, col3, col4) values (?, ?, ?, ?)

AQT will prepare this statement then, for each row being loaded, will supply the values for the parameters and execute the statement.

This method runs fast as the database will only need to interpret the statement once (during the prepare). The subsequent executes run fast.

Loading Using Inserts Statements

With this method, AQT an runs individual Insert statement for each row being loaded.

In some circumstances this is a preferrable method:

some databases do not process the prepare / execute method efficiently
some unusual data types can have trouble being passed through a parameter. Using Load Using Inserts often provides a more reliable method of loading.

You specify this load method by (on the Options tab) selecting Load using Insert statements.

Bulk Inserts

This is a variation of Load Using Inserts.

When this is specified, AQT loads multiple rows per Insert statement. Example:

insert into table (col1, col2, col3) values ('A',1,'a'), ('B',2,'b'), ('C',3,'c'), ('D',4,'d')

This statement loads 4 rows in a single operation.

This provides a faster method of loading than Load Using Inserts, however:

not all databases support this type of Insert statement
if there is any invalid data value or duplicate row then the Insert statement gets an error and none of the rows are loaded. This makes this option more suitable for loads which are likely to run without error.

You specify this load method by (on the Options tab) selecting Load using Inserts plus Use Bulk Inserts.

You can specify the number of rows AQT will load in a single Insert with Max insert rows.

Advanced Query Tool

https://www.querytool.com