Re: [PHPLIB] Re: [PHP3] persistent data From: nicholas klem (nklem <email protected>)
Date: 03/29/00

On Wed, 29 Mar 2000, Sascha Schumann wrote:
> On Wed, Mar 29, 2000 at 07:13:23PM +0200, nicholas klem wrote:
> > On Wed, 29 Mar 2000, you wrote:
> > > > Yes, but how is this persistence implemented? Is there a
> > > > mechanism for preserving these variables in server memory?
> > >
> > > Not in server memory. While server side session management
> > > would be awesome, I don't know how feasible that would be.
> > > As of right now, the only way I know of to keep persistant data
> > > is through the use of client side cookies.
> >
> > I'm experimenting with phplib, and I strongly suspect that it is possible to
> > configure the class sessions to store itself and children in some kind of
> > shared machine-memory. The example configuration file comes like this:
> >
> > #class Example_CT_Shm extends CT_Shm {
> > # var $max_sessions = 500; ## number of maximum sessions
> > # var $shm_key = 0x123754; ## unique shm identifier
> > # var $shm_size = 64000; ## size of segment
> > #}
>
> As the author of ct_shm I strongly discourage people from running it
> on production servers. The way shared memory is handled by the shared
> memory module is far from being efficient. For example, if you store
> 1000 variables and delete the first one, the module will copy 999 variables
> including the variable data to avoid memory fragmentation.
>
Thanks! A good thing to know. Maybe we could balance the load between mysql and
CT_Shm and see which breaks first? Anyone know of good ways to stress-test mysql
and/or php? Could be fun. ;-)

nicholas
nklem <email protected>
-
PHP3 Base Library Mailing List. Send messages to <phplib <email protected>>.
To unsubscribe, send "unsubscribe" to <phplib-request <email protected>> in
the body, not the subject, of your message.