Re: CORBA performance.

[Havoc Pennington <hp redhat com>]

Michael Meeks <michael peabody ximian com> writes:  
> 	Hmm; well you can do oneway calls which allow you to dispatch
> in a non-blocking fashion, you can also fairly easily create async
> stubs / skels that allow a larger degree of non-blocking-ness (
> although currently low level writes are synchronous which blows - but
> is convenient and fast for memory allocation in many cases ).

Well sure, but then you are no longer network transparent. ;-)
oneway/async are hacks that expose the details of networking.

The whole point I was making is that robust remote component usage has
to care about the fact that the component is remote.

For example, there is no reason GConf _can't_ be implemented with
CORBA - but really in the end the IDL _should_ have been something
like:

 interface ConfigServer {
   oneway void sendMessage (in string message);
 }
 
 interface ConfigClient {
   oneway void sendMessage (in string message);
 }

And CORBA is getting you exactly nowhere there - it's just being used
to avoid fooling with sockets by hand. This is just a traditional text
protocol, with no useful IDL and no useful transparency.

Havoc