in reply to
Realtime get with LWP or Mechanize
There is LWP::Coro which allows non-blocking LWP.
Could you elaborate why you want to do this as I would expect that usually the time spend on the application server for serving the http request will be longer than the time needed for network communication and the parsing will get more complex when you process incomplete messages - especially fif the parser needs to be really robust.
If you want to speed up things, look into threads with multiple LWP instances running requests in paralle.