[rabbitmq-discuss] RabbitMQ Clustering Million of Users

[Eugene Kirpichov]

Depends on the type of the requests and the usage pattern. E.g. if
you're fine with using non-zero prefetch, and the requests are just a
few hundred bytes or maybe a kilobyte, then 1 instance will be enough.

On Wed, Jul 4, 2012 at 2:54 PM, eleanor <evangeline.eleanor at gmail.com> wrote:
> Hi, I'm trying to determine how much rabbitmq nodes (in a cluster) would it
> take to handle 1 million of users. I'm trying to compare that to Google
> Appengine Instances. Let's say my application on GAE should process the
> following:
>   - 4000-5000 requests/second
>   - 400-500 instances running
>
> Now I'm not really an advanced user of GAE so I don't really know how to
> compare those results with rabbitmq.
>
> And I know you're not GAE users, so just ignore the first part of the
> message and answer me this:
> 1. How big cluster would I need to handle 4000-5000 requests/s?
> 2. Is it a good idea to rent the available servers on 6sync or some other
> VPS hosting to build a cluster of RabbitMQ nodes there or is something else
> really a better way of building a RabbitMQ cluster?
>
> That's it.
>
> Thank you
>
> --
> View this message in context: http://rabbitmq.1065348.n5.nabble.com/RabbitMQ-Clustering-Million-of-Users-tp20583.html
> Sent from the RabbitMQ mailing list archive at Nabble.com.
> _______________________________________________
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss



-- 
Eugene Kirpichov
http://www.linkedin.com/in/eugenekirpichov