Malcom, you're right in many points... I'm trying to identify the possible bottle-necks of the system, and hopefully, if we follow the "share nothing" architecture and the best practices I won't have many problems. The site is not that complex, the only point of danger could be a recommendation engine.
But, turning back to my question... Can you tell these sites with thousands or hundreds of thousands hits per minute? Ok, let's leave apart their stats... Any big name on the internet? slashdot? twitter is RoR... Thnaks!! On 23 abr, 11:51, Malcolm Tredinnick <[EMAIL PROTECTED]> wrote: > On Wed, 2008-04-23 at 02:13 -0700, bcurtu wrote: > > May you enumerate the largest django sites (apart from WSJ)? Do you > > know traffic stats of these sites? I'm building a big site, with > > potentially lots of traffic and I'm a bit afraid what I can expect... > > There are at least two problems with your question. Firstly, that type > of information is often commercially confidential. Secondly (and more > importantly for your case), "large" isn't a well-defined term. > > Performance depends on things like database size (the number of records > that are accessed to produce the pages), number of different sorts of > pages (which affects cachability), number of page views, number of > active users (if you're doing per-user customisation) and probably a > bunch of other things I can't think of off the top of my head. > > There are known examples of sites doing thousands of page views per > minute, other sites with users in the hundreds of thousands and possibly > more and yet other sites querying many gigabytes of data. At those > levels, you have to put some effort into performance, no matter what you > are using, but no problem is insurmountable. Keep an eye on performance > as you are developing, don't prematurely optimise, since many decisions > require the functionality to be known first, but do be prepared to try > out a few alternatives in various situations. > > If you have concrete performance problems, feel free to post a > description of the type of problem here. Best to have some measurements > to hand first, though, since optimisation is rarely productive when you > have to guess as to the significance of any particular feature. > > Regards, > Malcolm > > -- > A clear conscience is usually the sign of a bad > memory.http://www.pointy-stick.com/blog/ --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-users?hl=en -~----------~----~----~----~------~----~------~--~---
