Interesting if your planning on building a large scale web infrastructure 
of your own, less so if you plan on using some of the ~80% commodity 
hosting out there with PHP already installed.

It's a bit like me driving a 1985 Mk II Golf because it's easy to find 
cheap parts and someone saying a Nissan Skyline is better (because it's 
faster) while disregarding price of 
car/shipping/parts/insurance/fuel/labour/etc.

Either way, I think they missed a few key things (not that it would 
suddenly make PHP compete some of the top languages):

- They used the "old stable" PHP 5.3 instead of the faster "current stable" 
PHP 5.4 - maybe only 13% faster overall, but it seems to be able to sort 
arrays 45% faster (quicker than Python btw)

- They didn't enable APC or any other op-code caching - this can give you 
5x more requests per second (also, disable  `apc.stat` for another minor 
performance improvement)

- The raw PHP version uses a single database connection and a single 
prepared query in its loop, while the CakePHP benchmark doesn't even have 
persistent database connections enabled.

On Thursday, 28 March 2013 18:58:34 UTC, Miles J wrote:
>
> Found this article. Pretty interesting that Cake is last place in every 
> benchmark (given that it's multiple languages).
>
> http://www.techempower.com/blog/2013/03/28/framework-benchmarks/
>

-- 
Like Us on FaceBook https://www.facebook.com/CakePHP
Find us on Twitter http://twitter.com/CakePHP

--- 
You received this message because you are subscribed to the Google Groups 
"CakePHP" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/cake-php?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to