Spring JDBC vs JDBC

Question

I am trying to use spring 3.0 SimpleJdbcTemplate and it takes 5 minutes to insert 1500 records, while it takes me a few seconds. for insertion using direct JDBC. Not sure what I'm doing wrong.

Answer 1

If you are creating a batch package, use Spring batch - JdbcBatchItemWriter with the correct block size settings, which will load these 1500 records in less than a second.

Answer 2

Some things worth checking out:

• Overhead can be associated with a transaction managed by Spring at the application level. See which transaction manager you are using (look for a bean named transactionManager ). If you use JTA, probably where your problem is. Since this is fast with JDBC, the bottleneck does not seem to be db.
• Depending on how your application uses this transaction, it may contain everything in memory before it completes all 1,500 requests and completes the transaction. Do you see a big difference in memory usage (Spring needs to be much taller)?
• Which database connection pool do you use in any of the cases?

A quick way to profile your application:

Get pid - "jps -l"

Memory: jmap -histo PID (check if there is any form of memory leak)

Check what happens under the hood: jstack PID (look for slow or recursive method calls)

Answer 3

How about using

 jdbcTemplate.batchUpdate(new String[]{sql});