I am looking into various performance improvement approaches on our Java
process which process multiple database request (using JDBC) in sequential
order. We have one database table with column "status" which have possible
values say XX,YY,ZZ. So our Java process run in sequential order processing
all XX status requests first, then YY requests and then ZZ requests. Each
activity (XX,YY,ZZ) has been associated with multiple database connection,
database manipulation (insert, delete) and interfaces with other external
system. Main method is look like this:
main()
{
ProcessXX();
ProcessYY();
ProcessZZ();
}
I am planning to implement all these approaches:
1. Implement Connection Pooling in all these methods.
2. Define ProcessXX(), ProcessYY() and ProcessZZ() as a separate process and
run separately under NT "Services".
3. Within ProcessXX() method, right now I'm traversing each record via while
(resultset.next()). But if there are say, 10 XX requests at a time, then it
processes one-by-one. By this approach, 10th request will finish only after
previous 9 request finishes first. Is there any way so that all 10 XX
requests run separately so that all 10s finished in the same time?
Could you please guide me how can I design or implement #3 item. Any help on
this problem is greatly appreciated. Meanwhile I'll keep on looking into it.
Thanks.
_________________________________________________________________
Get a speedy connection with MSN Broadband.� Join now!
http://resourcecenter.msn.com/access/plans/freeactivation.asp
____________________________________________________
To change your JDJList options, please visit:
http://www.sys-con.com/java/list.cfm
Be respectful! Clean up your posts before replying
____________________________________________________
- [jdjlist] Re: Speed up process/Performance Improvemen... Parimal Amin
- [jdjlist] Re: Speed up process/Performance Impro... James A. N. Stauffer
