So far we have benchmarked situations where dataset > RAM or >> RAM. I think it is an interesting point to also analyze the case when dataset < RAM, to see how

So far we have benchmarked situations where dataset > RAM or >> RAM. I think it is an interesting point to also analyze the case when dataset < RAM, to see how efficiently both systems manage the caches, query planning etc. Stay tuned and thanks for the suggestion! :)