Hello, I'd try to perform some benchmarks of ASP.Net based CMS using mono but I got into few issues:
1. Performance stability issues. As more benchmarks I run this xsp/mod-mono/mono-server-fastcgi become slower. Even with simple application (like Hello World) I get 300 requests per second first time I run benchmark (ab -c 5 -n 500 URL) then this number become lower and lower and **never** returns to the original level. I've seen reports on same problem there as well: http://www.phpvs.net/2008/02/08/benchmarking-mono-aspnet-vs-php-a-slight-problem/ I also had written about it there: http://art-blog.no-ip.info/cppcms/blog/post/27 This makes all possible measurements irrelevant because the results are not consistent 2. I can't enable output cache support. I created a simple ASP.Net application that shows clock and added Output cache directive. that stores the cache for 10 seconds. I tested it under IIS refreshing the page under IIS does not change clock value. Same application caching hadn't given any result under xsp/mod_mono/fastcgi-server. And Yes, I had reas this FAQ: http://www.mono-project.com/FAQ:_ASP.NET#OutputCache_and_VarByParam I use ASP.Net 2 application. I run mono 1.9.1 --- under Debian 64 bit. I had tested all possible mono web servers (XSP2, apache-mod-mono and lighttpd+fastcgi setup). Question 1: Is there any way to solve above two issues? Question 2: Assuming that two above issues are solved, How mono web server is compared to IIS web server in terms of performance for **real world** application. Is it correct to assume that mono performance will be close to asp.net one. (I mean difference x2/x0.5 is fine for me but not order of magnitude). Thanks a lot, Artyom _______________________________________________ Mono-list maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-list
