[rabbitmq-discuss] Scaling RabbitMQ in "the cloud"

Tue May 6 18:54:07 BST 2014

I think the NYT description of their new Rabbit deployments is a great
place to start:

http://highscalability.com/blog/2014/1/13/nytimes-architecture-no-head-no-master-no-single-point-of-fa.html

-J

On Tue, May 6, 2014 at 10:31 AM, Ryan Brown <ryankbrown at gmail.com> wrote:

> I hate using that term because it is so over/misused. However, I am in the
> process of re-structuring my application to be deployed to ec2. My
> application wraps RabbitMQ with a friendly REST interface that does some
> additional routing and management for us. Right now, we have RabbitMQ in a
> four node active/active/active/active cluster behind F5 in our data center.
> What we need to do is push this out and have an instance of our application
> and a RabbitMQ cluster in each region of ec2 (if not each az). Ideally we
> would like to have the queues federated in such a manner that if a
> publisher of a given exists in New York City, that a subscriber in Sydney
> will get their message delivered from their nearest az. (This may be naive
> by me and par of what I am trying to determine.)
>
> What I would like to know is if anybody has experience doing this. Has any
> lessons-learned to share. Gotcha's? Maybe has something documented or knows
> of some documentation that exists that I am not finding.
>
> Best.
>
> Ryan
>
> _______________________________________________
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20140506/bafe34ba/attachment.html>