The core of the Artemis codebase is the HornetQ code that was donated to the Apache foundation several years ago, so both codebases have had their cores tested in production environments, and they've both had ongoing development and bug fixes, which always opens up the possibility of introducing new bugs. So I don't think the relevant distinction is between the maturity of the code bases, but what you're looking to get from the products, and what you can expect from their future development lifecycle.
In general, I would say to use Artemis unless you have a reason to use ActiveMQ 5, because as time goes on, more and more development will focus on Artemis and less on ActiveMQ 5. The one limitation of Artemis that I'm aware of is that all unconsumed messages must fit in memory in the broker, so if you are unable to work within that limitation, you would probably want to use ActiveMQ 5. Otherwise, I would recommend using Artemis, as it's the presumed future path for ActiveMQ development. Tim On Dec 7, 2018 1:33 PM, "Francesco Nigro" <nigro....@gmail.com> wrote: AFAIK Artemis has been used with success in production in many cases: I just think that is a matter of what you search in a broker. Given that I'm biased toward performance I know what Artemis can deliver from this pov and I can say that there is no really match (with many other brokers) related to this aspect :) Il giorno ven 7 dic 2018, 20:23 trevdyck <trevd...@amazon.com> ha scritto: > From what I understand Artemis is still quite new and has not been used in > many production environments yet. ActiveMQ has been hardened for many years > in production, so that may be something that factors into your decision. > > As for JMS2.0 you should look at whether 2.0 offers anything important that > you really need. > > > > -- > Sent from: > http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html >
