Just browsing the archives for info on the clustering strategy. At one
point I read the following:
> The drawback of this solution is that the client and server has to be
on
> the same subnet (because Jini uses multicast for lookup). This will
not
> be a big problem in most cases since the most common case is to use a
> webserver as client, which almost always is on the same subnet. If
> anyone objects to this assessment, please let me know and I will take
> into account clients on different subnets (Jini can work in this case
> too, but not as dynamic).
I would definitely take it into account. I even question the claim "the
most common case...". A very common deployment for security is to have
the web servers in a "DMZ" subnet, and app servers and database in a
protected net:
firewall ---> (DMZ) web servers ---> firewall ---> app server -->
databases
Sometimes the second firewall (between web and app) is eliminated and
basic subnetting is all that is used. In either case, multicast may not
be the way to go.
These deployments are usually used when security is a large concern. As
architects, we have to balance competing forces like flexibility,
manageability, security, availability, scalability, etc. As platform
"vendors", you have to decide what kind of deployments you want to
support. The kind I described here (which is common based on my
experience) may or may not matter to you.
As a friend of mine is fond of saying, "food for thought, eat what you
like"
-rg
