> 1. boolean setting and checking "yes" and "no" are faster than 1 and 0
I'd be curious to understand why this is so. Could it be that TRUE is really
held in memory as minus one (?) and so a tiny conversion is going on?
> 2. tens of thousands of lines of cfcomments really don't make a lick of
> difference
Thank god! What would Hal give lectures on if FuseDocs actually inhibited
execution performance. ? :)
> 3. sometimes client variables are faster than session variables (shhh -
> don't tell)
hmm. Well, without CFLOCK or with it? If SESSIONs can be faster than CLIENT
absent CFLOCK then does the test cover enough to kick in locking problems? A
lost SESSION variable can be as fast as it wants since it is otherwise
useless :)
> 4. application variables are slow as nuts (locked or not)
oh, damn, this explains a LOT of issues :) I got an app that is
APPLICATION-heavy, but only 5-10 users use it. This explains why it moves
like molases.
> 5. all this doesn't matter one lick if your hourly wage to implement all
> these ends up being more than the cost of another server
yes, and I'm laughing all the way to the Bank about it :)
keep up the good comments, this was some thought-inspiring performance stats
to keep on a back burner and thought about!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Structure your ColdFusion code with Fusebox. Get the official book at
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
