Github user franz1981 commented on the issue:
https://github.com/apache/activemq-artemis/pull/1361
@clebertsuconic Having only one instance would be optimal, but considering
that is a simple wrapper with no byte[] or off-heap memory allocated into it,
just avoiding 100K it's a good compromise IMHO.
About the changes on the NIO file factory, I've already tried (and
measured) a solution with the Netty pools before pushing that version, but I
found it to not be a good solution in this case for several reasons:
1. it is slower if compared to the TLAB slution I've implemented (the pool
is not shared across threads)
2, it grows the overall memory footprint and can be cause of leaks
3. it doesn't work nicely with "plain" ByteBuffers, forcing the use of
ByteBuf (not a good idea for normal file API)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
