Ok, I have a prototype running, or I should say crawling, to copy data from
an MS Access database over to an MySQL database.  I'd already set up the
tables and indeces, and was using a Java program to select all of the
contents from each table in the Access DB and insert each row from the
resulset into the MySQL.  The largest table has about 7700 rows and a couple
dozen columns,  It only takes a couple minutes to get the data from the
Access DB, but it seems to take hours to insert it into the MySQL,
undoubtedly because I've so far used the simplest approach: i.e. create a
prepared statement, and iterate through the resultset, populating the
parameters for the prepared statement and then executing it before
proceeding to the next.

I need a way to copy the data over that is much faster.  One options I can
see, but which appears to preclude using prepared statements, is to manually
parse the resultset obtained from Access to construct a really long SQL
statement (i.e. with each row consisting of comma separated values, enclosed
in parantheses, and separated from the next row by a comma).  There is a
question of how many rows ought to be inserted in one SQL statement.  A
second option suggested itself when I noticed that prepared statements have
an addBatch member function.  It wasn't clear how to use it, though, since
it doesn't appear to take any parameters, and yet the description says it
adds parameter values to a batch of SQL statements.

I do not know if either of these options I'm considering would improve
things.  Are there any other options I ought to consider?  I need to get
this done since I will need to deploy my application on a linux box.

Thanks,

Ted