> Without concurrent request dispatch, jruby's hypothetical greater thread parallelism is unlikely to effect requests-per-second benchmarks. Also, the benchmark

> Without concurrent request dispatch, jruby's hypothetical greater thread parallelism is unlikely to effect requests-per-second benchmarks. Also, the benchmarking test would have to be willing to issue requests concurrently! That's a good point. That said, JRuby would most likely benefit from a concurrent test, as the thread-per-request model of JRuby can scale much higher than the process-per-request model which is used a lot in the MRI world.