The version of PHP that you are using can change the peak memory usage considerably:
For example, I run several large applications all written in Laravel4 some of which have over 40 service providers and hundreds of models.
Another smaller personal application of mine running under PHP 5.5.9, peaks at 3.75MB after the op-cache is warmed up with each page loading in under 50ms all on a basic linode vps.
At the end of the day each of the different server environments serves a pretty extensive application in under 100ms, often under 60ms; however that doesn't matter as we stick them behind varnish meaning most page views get returned in under 30ms.
I am also fighting with performance.
On my local (MBP, mamp...) I am always having around 18MB and 500ms. On my AWS Elasticbeanstalk (m1.small) it's 700ms.... using APC on both hosts. I could save some ms by reducing some cache calls ( http://laravel.io/forum/02-28-2014-accessing-the-same-cache-key-multiple-times-a-request ) by putting the results in a temporary config setting for one request.
Actually my api is working ok with 100-200ms, processing almost the same information. Now I want to investigate my blade templates (which aren't that complex)
dont use APC but PHP native opcache
for me :
+1 for opcache!
This is really great, came down to 7MB and 200-300ms from 18mb and 700ms. I also installed apcu but not sure at the moment if it also helps.
Thanks!
Sign in to participate in this thread!
The Laravel portal for problem solving, knowledge sharing and community building.
The community