|
I'm asking about "perceived performance" - from a user's point of view, how do OSQA instances compare with Stack Exchange? There will obviously be difference because every OSQA instance is hosted on a completely different setting... but, are there are any general performance comparisons? How has your experience been so far? See also this question on meta-draw3cards. |
|
I don't think that is a reasonable comparison. SX is runs on software dedicated to that particular set of sites, on which all share the same functionality. There's no need for lots of options, multi db support, i18n support, module support, etc, etc. Even themes are css only apparently. Only thing I can say is that OSQA behaves fairly good under pressure, but there are several variables you need to consider, when trying to build an high performance OSQA. Type of server, mod_wsgi vs mos_python, how do you serve your media, type of cache being used if any at all, etc, etc. My experience with my own OSQA installation is a bit less good than meta.osqa.net. Perhaps the fact that I didn't yet migrate to a real database (from SqlLite) is the reason... Will do it first chance I get. See also: http://meta.osqa.net/questions/6180/osqa-performance-tips
(16 Dec '10, 20:38)
ripper234 ♦
|
Do we really need yet another variation on this theme, @ripper234? Aren't there enough of these already?
@rickross, I understand you're probably sick of these by now, but I'm genuinely curious about this. Please check out the answer to the questions I posted on meta-draw3cards, the first user of our meta site finds it definitely slower.
See also: http://meta.osqa.net/questions/6180/osqa-performance-tips