RFR: 8287904: Reduce runtime of java.net microbenchmarks

[Claes Redestad]

On Thu, 16 Jun 2022 13:51:34 GMT, Jorn Vernee <jvernee at openjdk.org> wrote:

> Why the deleted ones?

When I evaluated those they appear to have excessively high run-to-run variance. Sometimes varying integer multiples within the same fork, and more warmup doesn't do much to improve it. There's also some overlap with other benchmarks. Removing reduces coverage marginally, but a benchmark that just produces noise is less than useful.

-------------

PR: https://git.openjdk.org/jdk/pull/9182