Np Michael, I understand it at all :) It is the type of feedback besides perf ones that I think would help: I was only concerned to know if any evident regression was already perceived and I 'm happy it is not (correct me if I'm wrong!).
Given that I'm 100% of the idea of having it configurable, to revert to the fixed base approach ,but I can push it a little further: I prefer to think that this configuration is not a matter of old Vs new, but a first class citizen option. Because is not fair to consider the new one as "experimental": it is not, in the strict sense, is simply different and new. I'm sure that giving a user (dev in this case) the power to choose between different solutions is always a good idea :) The rest will be natural selection: if an option won't be used anymore ,will be deprecated and so on...makes sense? -- View this message in context: http://activemq.2283324.n4.nabble.com/DISCUSS-Artemis-IOPS-Limiter-strategy-tp4725875p4726123.html Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
