Connection Pooling
-
Please Help....I've been given the following problem, and I'm trying to get some information about technology/techniquest to solve it: 1. System model uses a dbms with a limited number of connections avaiable. Therefore, all communications to the dbms must go through a shared connection pool. 2. Various processes within the system will need to execute queries against the dbms through a pooled connection and then operate directly on the recordsets. Now, the last time I did a connection pool to a dbms it was a set of classes that were compiled into a process and used ODBC to make the connections to the dbms. All updates to the dbms went through this single process. What's confusing me is that some developers here are trying to convince me that I can make a connection pool in one central-process, then pass these connections out to other processes that will "do work" and then release the connection back into the free pool held in the central-process. I just don't see how this can work.....Can anyone help point me in a direction where this kind of thing has been done before? Thankx!!!! Just trying to keep the forces of entropy at bay