evaluating clustering options for AMQ 5.9.1

Hi I am doing some R regarding clustering for AMQ 5.9.1. We are looking to
evaluate active/active options as well as active/passive options.
Master/slave pattern seems to be popular for failover.

I've looked at the following blogs:

http://bsnyderblog.blogspot.com/2010/01/how-to-use-automatic-failover-in.html
<http://bsnyderblog.blogspot.com/2010/01/how-to-use-automatic-failover-in.html>

http://sensatic.net/activemq/new-activemq-failover-and-clustering-goodies.html
<http://sensatic.net/activemq/new-activemq-failover-and-clustering-goodies.html>

http://tmielke.blogspot.com/2011/09/activemq-network-bridge-to-masterslave.html
<http://tmielke.blogspot.com/2011/09/activemq-network-bridge-to-masterslave.html>

Please advise what would be the best choice for us considering active/active
scenario. Network of brokers? If so, what is the best example/tutorial
that I can deploy to a dev lab with VMs? thx.