Re: [Vrs-development] VRS and SEE/DEE goals

[Tim Terlegård]

> You're going along the same thinking as me :o)

Great!

> I would modify your 'thinking' slightly and say that:
> "Every VRS node runs a lookup service, which enables anyone to browse the
> web services available on the complete VRS.  Because there may be multiple
> LDS's acting as public entry points to the VRS cluster, each LDS must
> advertise the same services in their lookup service."

This is exactly what I was thinking of too  :-)
So when an LDS gets a web service added, it tells every peer in the VRS 
network about this? Otherwise they won't find out about the new service? This 
will cause more network traffic than just having a central Service Discovery 
Server to which services are added, but will be more robust as maybe all the 
thousands of VRS nodes know what services are available (without asking 
someone else).


> Any service deployed within the VRS may be tagged as 'private'.  These
> servicdes will NOT be advertised to the outside world by public facing
> LDS's.
>
> > This allows a VRS cluster to have a single server as a lookup directory
> > (although this makes it less p2p, the webservices can still be requested
> > in a pure p2p manner.).
>
> Hmm.  Multiple lookup servers would be good.
> This is kind of extended DNS isn't it.

Didn't you say you wanted every peer to know of all the webservices? Then 
central lookup servers wouldn't be necessary.

Maybe Lookup Services (or Service Discovery Server) suit better on central 
servers where anyone can ask what services are available (no information 
about who provides the services are exposed though). VRS is a p2p "thing" 
(didn't know what to call it  :), but everything might not have to be p2p?

-- Tim