Re: [Vrs-development] Scalability

[Chris Smith]

On Friday 26 July 2002 21:09, Ian wrote:

> well. what about something like this. we can treat all message routing
> on demand. by message, i dont mean every packet. let's treat it as a
> truly decentralized p2p network. for example, discovery of services
> would be done dynamically. 

I've not read the rest of this post yet.... but could you clear something up 
in your general terminology.

We need to differentiate between services.

We've got 'middleware' services, which are just application functions 
deployed within the middleware application space.

And we've got, erm, for want of a better description, webServices which are 
user services deployed into the VRS by LDS owners.


The Goldwater Middleware of course needs to keep track of what 'middleware' 
services are available, where they are and the expense of calling each, 
alternative locations (and the expense) etc.  There is a lot of fundamental 
dialogue which goes on between domain nodes which cannot be replicated 
through discovery type routing.  Also these 'service' requests across domains 
etc need to be (and will be eventually) subject to transaction processing.
Ie you begin a transaction, make several middleware calls (which may go to 
any arbitary domain), then rollback or commit the transaction.  You're not 
going to get this level of control, security and integrity from dynamic 
discovery aproaches.

But that is if you were talking about middleware 'function' services..... if 
you were talking about VRS deployed user services then that's totally 
different.


I'll read the rest of your post now, but I think we need to be clear from now 
on.

I've been using the term GWService, GWDomain, GW* etc to qualify what I 
mean... ie when talking about anything goldwater related, prefixing it with 
gw.

Chris

-- 
Chris Smith
  Technical Architect - netFluid Technology Ltd.
  "Internet Technologies, Distributed Systems and Tuxedo Consultancy"
  E: address@hidden  W: http://www.nfluid.co.uk