On Aug 30, 12:41 am, "Dr. Tarique Sani" <[EMAIL PROTECTED]> wrote:
> On Sat, Aug 30, 2008 at 1:17 AM, Gwoo <[EMAIL PROTECTED]> wrote:
>
> > Since we have spent some time over the last week optimizing, I figured
> > I would run some more tests. I really don't like "ab" as a utility. I
>
> But wha
On Sat, Aug 30, 2008 at 1:17 AM, Gwoo <[EMAIL PROTECTED]> wrote:
>
> Since we have spent some time over the last week optimizing, I figured
> I would run some more tests. I really don't like "ab" as a utility. I
But what was the test?
Can we have a peek at the areas where Cake 1.2 is being look
Boo, premature optimizers!
Hooray, beer..
On Aug 29, 3:47 pm, Gwoo <[EMAIL PROTECTED]> wrote:
> Since we have spent some time over the last week optimizing, I figured
> I would run some more tests. I really don't like "ab" as a utility. I
> have trouble believing that it accurately tells us what
Since we have spent some time over the last week optimizing, I figured
I would run some more tests. I really don't like "ab" as a utility. I
have trouble believing that it accurately tells us what is happening.
For one, I think mod_rewrite has an effect on the stats. IF someone
can prove otherwise
I always find it interesting how benchmarking is done.
My understanding of software development is to define some
requirements, write some test, run the test, write code to support the
test, once all test for given requirements iteration are passing, then
we look at profiling.
In all my years of w
.. and cake hello world application memory usage is 5 Mb.
So when you put few things more (model recursive instances) you must
fight to be below PHP4 default 8Mb.
(For exmpl. CI has 1.2 Mb.)
(persistModel and which caches class registry use even more memory)
(profiling cake I can see around 2000-
@kiang: That's a great idea. I think a more general performance-
tuning guide would really help out. I know there are some resources
on this, but I think a comprehensive guide in the Cookbook somewhere
would be great. I'll see what we can do about this.
On Aug 28, 11:11 am, kiang <[EMAIL PROTE
Just my two cents: CakePHP is fast enough for me :-)
That's not to say you shouldn't optimize for speed at some point in
the future. I just mean, it shouldn't be a priority.
Keep doing what your doing Cake dev's. Cake hasn't become one of the
most popular PHP frameworks because of its speed ;-)
Maybe there could be one page like 'Read first before benchmarking' in
home page since there exists many mis-understandings around the
world. :)
---
kiang
On Aug 28, 10:41 pm, Gwoo <[EMAIL PROTECTED]> wrote:
> One thing we know for sure is that we can always work to improve the
> speed. However,
> One thing we know for sure is that we can always work to improve the
speed. However, we will not do it in earnest until the time is right.
We have been here before, seen similar things, and come up with the
answers.
This is SO smart and the way leaders lead.
I am in the middle of making a HUGE
One thing we know for sure is that we can always work to improve the
speed. However, we will not do it in earnest until the time is right.
We have been here before, seen similar things, and come up with the
answers.
While the hello world is pretty silly, a couple of things strike me
right off the
Sorry, but Rasmus' benchmarks are useless. There's no documentation
of his methodology. There's not even any indication that he wasn't
running in debug mode. I've spoken to Paul M. Jones (author of the
original PHP framework benchmarks), and he's getting ready to run
another round soon. So, we'
On Thu, Aug 28, 2008 at 6:03 PM, kiang <[EMAIL PROTECTED]> wrote:
>
> http://talks.php.net/show/froscon08/
>
> He compared the performance of several PHP frameworks and CakePHP
> seems to be the one which have the biggest space to improve. ^^||
H Another "Hello world" benchmark for frame
http://talks.php.net/show/froscon08/
He compared the performance of several PHP frameworks and CakePHP
seems to be the one which have the biggest space to improve. ^^||
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
14 matches
Mail list logo
