Miguel, I will try to build a self contained sample. However, my tests are based on a real app, therefore it might be time consumming to do so. After testing other applications that are also using sockets extensively, my feeling is that mono is quite performant with sockets but less with byte array allocation / manipulation (however, this is not a proven observation).
Gonzalo, I'm using an open source library in this app (namely qpid client library). I'm 100% sure that this library could be optimised in several ways, but that's not what I'm trying to achieve. My goal is to compare .Net/mono with real apps and not optimising the various pieces. However, I still followed your advice and hacked quickly into the library to replace sockets with async sockets. As far as I can tell, it has no significant impact on the performances (both on windows and linux) Thanks to both of you for your advices, Cheers, Julien -- View this message in context: http://mono.1490590.n4.nabble.com/Mono-and-IO-Performance-tp3412393p3417187.html Sent from the Mono - General mailing list archive at Nabble.com. _______________________________________________ Mono-list maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-list
