Comments on: Benchmarking PHP HTTP Clients http://matthewturland.com/2008/11/23/benchmarking-php-http-clients/ Fri, 11 May 2012 20:03:11 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Matthew Turlandhttp://matthewturland.com/2008/11/23/benchmarking-php-http-clients/comment-page-1/#comment-234 Matthew Turland Tue, 25 Nov 2008 09:03:09 +0000 #comment-234 As far as I can tell, Ubuntu does not have a DNS cache running by default. nscd is available as a package for it and I tried installing it, but found this comment in the configuration file:# hosts caching is broken with gethostby* calls, hence is now disabled # per default. See /usr/share/doc/nscd/NEWS.Debian.So I uninstalled that and followed this tutorial to install, configure, and test dnsmasq instead:http://ubuntu.wordpress.com/2006/08/02/local-dns-cache-for-faster-browsing/I also amended a few oversights in my original example. The curl options array should include CURLOPT_FOLLOWLOCATION and the pecl_http options array should include 'redirect.' Using a GET request that results in a redirect as the test subject, I got the following results.Before installing dnsmasq: pecl_http - 12.75% streams - 25.27% curl - 12.92% pear - 32.82% zend - 24.50%After installing dnsmasq: pecl_http - 19.58% streams - 21.40% curl - 17.30% pear - 28.20% zend - 21.10%So, a local OS-level DNS cache does bring the performance of these clients significantly closer to equal footing. Lars and Pierre, thanks for helping to clear this up! As far as I can tell, Ubuntu does not have a DNS cache running by default. nscd is available as a package for it and I tried installing it, but found this comment in the configuration file:

# hosts caching is broken with gethostby* calls, hence is now disabled
# per default. See /usr/share/doc/nscd/NEWS.Debian.

So I uninstalled that and followed this tutorial to install, configure, and test dnsmasq instead:

http://ubuntu.wordpress.com/2006/08/02/local-dns-cache-for-faster-browsing/

I also amended a few oversights in my original example. The curl options array should include CURLOPT_FOLLOWLOCATION and the pecl_http options array should include ‘redirect.’ Using a GET request that results in a redirect as the test subject, I got the following results.

Before installing dnsmasq:
pecl_http – 12.75%
streams – 25.27%
curl – 12.92%
pear – 32.82%
zend – 24.50%

After installing dnsmasq:
pecl_http – 19.58%
streams – 21.40%
curl – 17.30%
pear – 28.20%
zend – 21.10%

So, a local OS-level DNS cache does bring the performance of these clients significantly closer to equal footing. Lars and Pierre, thanks for helping to clear this up!

]]> By: Lars Strojnyhttp://matthewturland.com/2008/11/23/benchmarking-php-http-clients/comment-page-1/#comment-233 Lars Strojny Mon, 24 Nov 2008 14:04:51 +0000 #comment-233 Was nscd (name service cache daemon) running? Don't know what's the default on Ubuntu. Was nscd (name service cache daemon) running? Don’t know what’s the default on Ubuntu.

]]> By: Pierrehttp://matthewturland.com/2008/11/23/benchmarking-php-http-clients/comment-page-1/#comment-232 Pierre Sun, 23 Nov 2008 19:13:26 +0000 #comment-232 CURL uses its own DNS cache system, independently of OS (if it implements any). The PECL HTTP uses CURL and its DNS cache (via the option dns_cache_timeout).I do not think php has a DNS cache (but only did a quick grep, I'd to check again more deeply. CURL uses its own DNS cache system, independently of OS (if it implements any). The PECL HTTP uses CURL and its DNS cache (via the option dns_cache_timeout).

I do not think php has a DNS cache (but only did a quick grep, I’d to check again more deeply.

]]>