Hi Wim and welcome to the Netduino community.
A micro platform will always have more limitations then PCs. To be honest, I would never even think about making a Netduino a server. My approach for communicating with a Netduino and the web is set up differently:
I've got a webserver, with Linux, Apache, MySQL, PHP. On this server I run a small socket listener with MySQL support as well.
My Netduino connects to the socket and has an open connection all the time with the webserver. All requests are translated to a protocol that only contains data. No markup, no image files, etc. They're all stored on the webserver itself.
That method is actually very fast and makes communication with the netduino instant, without losing much resources on the Netduino.
I know this doesn't answer your question: "how to speed up the webserver", but it can help you with a faster approach.
Currently my socket listener is very beta (if not alpha), so I'm not going to share code in it's current state. But it's a project I'm working on.
Hi Stefan,
I'm not running a complete webserver on a Netduino. It's just a tiny web application to setup an intelligent device. Like you will find a lot today in network printers, routers, etc.. So a LAMP stack (Linux, Apache, MySQL, PHP) is out of the question. The largest amount of data it will ever have to send is between 10 and 20k. That's not a lot, but it takes 10 to 20 seconds. And that's way to long, I think, for a 32Mhz processor. I used to have an 8Mhz Archimedes computer (whith the first ARM processor) in the early eighties which was way faster. I just can't understand where the cycles go to.
Anyway, I'm considering a Raspberry Pi now to do the job. An amazing $25 creditcard sized Linux computer with a 700Mhz ARM and 256MB of RAM.
Regards,
Wim Roeling