Comparative Benchmark in test of load of Apache, Zope, SPIP, Drupal, PHP-Nuke and Templeet, on OpenBrick.   By Rafael Pinilla, April 20, 2003


Read here... (Google Translation)



Original page: here...





H.S.

Comments

kika’s picture

babelfish translation:

The graphs speak alas about them even. Drupal manages to make less well that PHP-nuke, with the favour of a mechanism of mask based on a requ

dries’s picture

I think so, yes. I suspect caching was either disabled or not working properly (Did thhe benchmark program send the proper headers?). Also, it is not clear what modules have been enabled (Were all modules enabled? Was the locale module enabled?).

Hutchison@drupal.org’s picture

I came across this article independently as I still struggle between choosing a CMS. I've been digging into Drupal but keep getting torn by Xoops, Zope, etc.

One key benefit for me is my perception that Drupal is pretty fast. I'm wondering if anyone has done some benchmarking with Drupal 4.2 or newer, with caching enabled.

killes@www.drop.org’s picture

Drupal has received many improvements over the last months so that test is only of historical interest.

There were some benchmarks posted recently but they only compared Drupal and a patched version of Drupal.

Steven’s picture

Very weird results indeed... the test uses 4.1, which does not include the recent bootstrap changes, but even then I can't imagine those results being correct.

Drupal.org and Kerneltrap.org have both endured Slashdotting several/many times without any noticable slowdowns, and with just a standard codebase. According to these results, Drupal would have gone up into flames after a couple of minutes.

dries’s picture

I suspect that Drupal was not configured properly. Drupal's caching feature is disabled by default so maybe they forgot to turn it on?