oak-run benchmarks

[Julian Reschke <ju...@gmx.de>]

Hi there,

I'm currently looking at the benchmark behavior for the RDB persistence, 
and I believe I'm seeing degrading performance with each additional run 
of the benchmark.

To make cases like these easier to find, would it make sense to also 
report on whether there's a trend in benchmark times? (such as by 
reporting the average ratio of runtime between subsequent runs?)

Best regards, Julian