Does anyone know what the relative performance would be of using updateable
result sets versus using individual insert or update statements?
(Specifically at the moment I'm dealing with JSP programs using J/Connector
on Tomcat, if that matters.) For example, if I want to read ten records and
update them, would I be better off to use an updateable result set and
update each as I go, or to issue UPDATE statements for each? (Intuitively it
seems to me that the ResultSet should be faster, but who knows?) Similarly,
if I want to insert 10 records, would I be better off to issue 10 INSERT
calls, or to create an empty Result Set and then add records to it?
Another efficiency question: I'm working on a JSP program where I build a
display based on a select, the user fills in some fields that relate to each
record, and then when Submit is pressed I re-read the records and do my
processing. I was just assuming because it sounded right to me that I am
better off to SELECT all the records, get the result set back, and then loop
through, matching the records read against the fields off the web form, and
whenever I see that I have to do something, I go off and do it (updating
various tables based on input). But it occurs to me that an obvious
alternative, and easier to code, would have been to loop through the web
form fields, for each do a select that retrieves that one record, and then
process it. That is, it seems to me intuitively that I am better off to do
one select that retrieves 100 records and loop through the result set
processing them, rather than to do 100 queries that each retrieve 1 record.
Is this true?
Thanks for any help!
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/my***********...ie.nctu.edu.tw
