Just to chip in on the subject of benchmarking as well. I'm not a huge fan of apache ab. I've run into issues with it in the past when testing php based sites. Results like those indicating that a framework can outperform a simple php script make me question its truthfulness. I find that siege gives more consistent and realistic results.
Its great to see that people are concerned about the performance that cake has. Its something that I and others on the core team have spent a fair amount of improving when working on 1.3. I'm glad to see see that others were able to reproduce the speed improvements I saw when, I last ran benchmarks as well :) -Mark On Mar 8, 4:45 am, WebbedIT <p...@webbedit.co.uk> wrote: > > Gee! Now I am really confused... Ferrari 599 is 1.2.6 and Prius is > > 1.3.0-RC1? > > Lol, not quite, I compared your test to testing the newest model of a > Ferrari against the outgoing model but using the criteria that we > would use to test the Toyota. Made sense to me :) > > > Anyway, the whole idea of hello world benchmarks is to assess the > > frameworks weight. > > But you cannot assess the weight of a framework if the test doesn't t > stress the majority of that framework. > > > Even though you and I do not agree on this methodology, I do bet you > > want too a lean framework capable to deliver as many requests/second > > as possible. > > I would be very interested to look at similar benchmarking on a large > app. Hopefully I will be in a position to do this later in the year > as I am currently developing a multi-subdomain sector dedicated portal > which should get a lot of traffic in Cake 1.2.6 and once I have the > money to hire a development team I will be asking them to port the > site to 1.3 and compare the two. > > > Should you think there is a better way to assess this measure, please > > tell me, I am glad to know of it. > > I don't know of a better way as I'm yet to develop or run a site where > it's performance has been called into question. I can tell you that a > client of mine who moved from a large bespoke CMS I coded (before I > started using Cake) to Drupal has had a nightmare ever since. > > But the two systems were never comparable and I very much doubt the > client will pay me to replicate his existing site with Cake and > compare the performance of the two (I'd bet a lot of money on Cake > performing better) but even if I could do this how would I make the > test a fair comparison? > > For now myself and all my clients are very happy with cake 1.2.x > performance and from your test it seems 1.3 is better so hoorah! Check out the new CakePHP Questions site http://cakeqs.org and help others with their CakePHP related questions. You received this message because you are subscribed to the Google Groups "CakePHP" group. To post to this group, send email to cake-php@googlegroups.com To unsubscribe from this group, send email to cake-php+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/cake-php?hl=en
