Hi, My application environment is the following: 1) The same application is duplicated on multiple servers. 2) These servers are loadbalanced by hardware in front of the servers which basically round-robins between the servers. 3) All client requests via http are session- and stateless. 4) Each server has the same CMP2 beans as all the servers, referencing data from a single shared database. 5) The number of servers can be large. 6) No JBoss clustering is used, due to lack of knowledge and perhaps 5 above. 7) Some CMP2 beans will write/change data to the database. 8) The main table to which most other beans relate will have about 10-20 Million records.
My questions are the following: How do 1) ensure that data across all servers is correct and valid. 2) ensure that CMP2 works fast and efficient taking 1) into account. Can you do 1) only by choosing a commit option that reads from database every time. If so, will performance suck? Also, some people reckon that a way to get rid of the CMP2 problem is by using something like Toplink from Oracle. I guess it is similar to Hibernate. I do not see how this will change anything. Any comments? Kind regards, -- Alwyn Schoeman SMART Money Inc. If you're a SMARTY and you're using the better DNS connections, you can access my homepage at http://nevyn.smarties.com.ph/Members/alwyn. The other unlucky ones: http://10.126.68.24/myplone ------------------------------------------------------- This SF. Net email is sponsored by: GoToMyPC GoToMyPC is the fast, easy and secure way to access your computer from any Web browser or wireless device. Click here to Try it Free! https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl _______________________________________________ JBoss-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-user
