Nils Bandener

@cwperks > What work is remaining on this PR in order to flip it from Draft? Mainly reviewing test failures and reviewing the special cases, like migration and API -...

@willyborankin @cwperks I did one more minor commit: https://github.com/opensearch-project/security/pull/4546/commits/5ebe5007c4a1cf64b1ed1129d0c3b1c7d9cdf11b ... could you please re-approve?

@cwperks > If its not possible to create a `static` roles mapping, then I think the test scenario can be removed. Good point, removing the part!

@cwperks > My understanding is that reserved roles can be assigned, but that the test is pertaining to `roles_mappings` Yes, it operate mostly on role mapping docs, but it is...

Please have a look at the approaches. I would be very interested in your opinions. As mentioned above, the implementation is not complete yet. The code contains a couple of...

I have also started to work on the micro benchmarks as discussed in #3903. The generally accepted standard for micro benchmarks in Java is the JMH framework. However, this is...

@cwperks @peternied @DarshitChanpura @scrawfor99 Just FYI: I worked a bit on the micro benchmarking part of this issue. As JMH was out due to its license, I reviewed other frameworks....

@cwperks > Is there any work remaining before labeling this PR ready for review? Yes. Actually, I was also about to ping you regarding this. - The heap used by...

# Benchmark results After having shared the micro benchmarks before, I can now share results of benchmarks performed on an actual cluster. This brings the following benefits: - It puts...

Side note: This is not the end of the story - there's still significant potential for performance improvements. I will file issues about that soon.