Michael Haggerty
mhagger at alum.mit.edu
Wed May 2 19:35:31 EST 2001
"Bryon G" <db8coach at hotmail.com> writes: > The syntax for the AOLServer database access stuff was, in my > opinion, harder to learn than the pg way largely because there was > less documentation available, but once that was out of the way the > pooling question seemed to be the only real difference between the > two in terms of end performance. I'm hope nobody is offended that I follow up with a comment about PyWX on this mailing list... With PyWX you don't have to limit yourself to the AOLserver database API to get connection caching. (In fact, I don't use it myself.) The latest version of PyWX has a single-interpreter, multiple-thread mode with which you could open, for example, multiple pg database connections on startup and store them in a pool. Individual scripts would grab a connection from the pool, use it, clear it out, and return it to the pool to be reused. Of course this is predicated on pg and your pool code being thread-safe. Once multithreaded Apache is out, similar tricks should be possible in the Apache world. Michael -- Michael Haggerty mhagger at alum.mit.edu
|