Re: [Vrs-development] Scalability

[Chris Smith]

On Monday 29 July 2002 16:01, you wrote:
> > 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.
>
> Yes, agreed. let's call them middleware or internode services. those are
> basically services provided to other LDS's from another LDS. I have no
> problem calling the other type webservices, or we can just call them
> EOD's (executable object datasets). All webservices are going to be in
> the form of EOD's right?

Ah!  But one LDS calling a 'service' at another LDS does so by just calling a 
middleware service.  The middleware itself knows that it is not local and 
breaks out onto the network automatically.  The 'application' that has just 
made the call to the middleware service has no idea that a network has been 
traversed.  Of course, an application may 'qualify' the domain that it wants 
to satisfy its service call if it wishes, and at that level the distributed 
nature of the application is known - but it all feels like a single unified 
application.

> > You're not
> > going to get this level of control, security and integrity from dynamic
> > discovery aproaches.
>
> well. hmmm. that is what i've been doing last year as research. this is
> really bleeding edge in computer science (ie not very complete). all the
> things we take for granted in a traditional environment is not there in
> a distributed environment. you know what though, all the dynamic
> discovery of webservices and stuff will never be in the initial rollout
> of VRS. so it doesnt matter. i dont want to get into what i think about
> this and get everyone confused because it is very out of context right
> now. let's just meet (me you and eric) and talk.

And this is where terminology is important.  Note you said 'discovery of 
webservices', and if we're talking about 'wbservices' then that's fine.
I just want to make sure that you're aware that messages generated by one LDS 
(like 'please give me a block of data from your RM') destined for another LDS 
are just middleware service calls, the fact that a network is crossed is, 
well, hidden.  The same call is made whether the middleware service required 
is provided by the local LDS or a remote LDS.

Clients and other 'external' agents requiring resources from the VRS (ie 
webservices) do so through a network server interface....  discovery here is 
not a problem.

Anyway, that's why I wanted to be clear about terminology, otherwise I'll be 
talking about middleware service calls between LDS's and you'll be talking 
about 'webservicesque' calls between LDS's and neither of us would spot the 
mistake until we'd got no where with the discussion for a month! :o)

Of course there is no reason why both can't co-exist, which is the point of 
our lovely design discussions!  We're obviously hitting this from different 
directions and talking about different things which is adding to the 
confusion.


Oh, see a post I've made about Goldwater Domains...

Cheers,
Chris


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