Hi to everyone! I'm searching feedbacks about a new implementation of org.apache.activemq.artemis.core.io.SequenceFile that provides (a lot) faster performances of both the "plain" Nio and libaio implementations: in "pure Java" (*libaio is not available for windows* AFAIK)! This implementation could open the doors to other improvements, impossible ( aka hardly possible) with the other implementations (e.g. IPC tools to receive/monitor/collect/visualize/transform the added journal entries with a sub-microsecond latency and provides some kind of analytics on them). The current implementation is a good proof-of-concept and anyone interested could try the bench attached to it:
https://issues.apache.org/jira/browse/ARTEMIS-508 <https://issues.apache.org/jira/browse/ARTEMIS-508> As soon as I'll proceed with the implementation i'll post updates. Anyone have comments, ideas, suggestions? Regards -- View this message in context: http://activemq.2283324.n4.nabble.com/Sequential-File-Improvement-Performance-Tests-tp4711306.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
