[openamq-dev] round-robin queues

[Esteve Fernandez]

El Sábado 09 Febrero 2008 14:11:04 Martin Sustrik escribió:
> Esteve,
>
> This is interesting. We haven't seen a scenario that wouldn't fit AMQP
> model so far. May you share your use case with us? If your point of
> "this kind of problem can be solved only be creating > 1,000,000 queues"
> is strong enough it may even mean there is a problem in AMQP protocol
> itself.

Although this makes me feel rather special :-) probably there's already a way 
of solving this in the current AMQP model that I'm not aware of.

Following the library use case, the idea is that each box has a database that 
contains the complete set of a specific type, and there's an application in 
some other machine that requests elements that are either magazines, books or 
dvds. Some types are requested more often than others, so they're replicated 
across several boxes (that's why I would like to do round-robin at this 
level). A box could hold more than one type, but given two boxes that contain 
the same type, they might not share the same database with the same types, 
e.g. A and B boxes could contain the book type, but B contains the dvd type 
as well.

The problem is that I'm storing millions of types (not just three or four), 
and they're independent, a magazine is not a subclass of the book type, for 
example.

As I said before, I'm an AMQP newbie, so feel free to criticize my model :-) I 
know there must be a better way of doing this rather than creating one 
million queues or adding an "x-distribution: round-robin" argument when 
declaring an exchange (and hacking it into OpenAMQ) that I've missed.

Cheers.