We got this question a lot - and our answer has been yes for more than three years. 6% of all clusters at CloudAMQP has HiPE enabled - and we are today running thousands of nodes. We have not seen issues with HiPE since 2015.
Erlang VM segfaults was a common issue before, and therefore HiPE was marked as experimental in the RabbitMQ's documentation. Just make sure that you are running a reasonably new Erlang version before you enable HiPE.
HiPE increases server throughput at the cost of increased startup time. When you enable HiPE, RabbitMQ is compiled at startup. The throughput increases with 20-80%. The drawback of HiPE is that the startup time increases quite a lot too, about 1-3 minutes. It’s therefore not recommended if you require high availability; you want your cluster to come back quickly in the event of failure.
It's possible to enable HiPE, High-performance Erlang, for your dedicated RabbitMQ clusters via the CloudAMQP control panel.
If you have any questions, you can reach out to firstname.lastname@example.org.