Re: PHP web farms (i.e. why msession or something like
| From: | mlwmohawk at mohawksoft dot com | Date: | Thu, 30 May 2002 02:49:14 +0000 |
| Subject: | Re: PHP web farms (i.e. why msession or something like | ||
| References: | 1 | Groups: | php.dev |
| Request: | Send a blank email to php-dev+get-84559@lists.php.net to get a copy of this message | ||
> mlwmohawk@mohawksoft.com wrote:
>> MySQL's table locking during update pretty much make sure that
>> performance will degrade under heavy load with many connections.
>>
> It's not actually as bad as you think. With a table like this, the
> table locking actually has a very minimal effect on it.
>
> In any case, I'm not arguing that using a database would perform better
> than using msession. I'm simply arguing that it would be good to have
> the built-in session management support databases. It might also be
> advantageous to add msession support to the core session support.
I think my position on msession is misunderstood. Of course *I* like it because *I* wrote it,
however, if I
had it to do over again, I probably wouldn't do it the way I did, and I would have more
reservations about
submiting it to the group. Participation in this group has certainly been a mixed bag.
When all is said and done, the reason I wrote msession still exists. There is a need for something
which
links multiple PHP systems together. Databases incur a disk I/O, the same as the file based session
module,
so maximums are controled by the disk subsystem. Msession is most similar to the mod_mm session
manager, it
has no disk I/O, although it does have network I/O latency.
I don't know what *the* answer is. I have contributed msession as *an* answer.
Idealy, there would be a "mod_net" which performs similarly to the session extension
portion of msession.