[hox]

Not only that, but there is no attempt made to gather metrics based on similar configurations / feature sets. The author specifically mentions that the AMQP systems persist their messages to disk by default, and this was the configuration used for the testing. How, then, are the "benchmarks" even comparable to the ephemeral message queues that don't provide any sort of persistence? Why wasn't persistence turned off to provide more comparable tests?

And why were nanomgs / 0mq even included?

I despise articles like this. They only serve to clutter up useful communication on such technologies.

[tylertreat]

Persistence was disabled. Nano and ZMQ are in their own group and aren't even attempted to be compared to other groups.