h2oai / db-benchmark

reproducible benchmark of database-like ops
https://h2oai.github.io/db-benchmark
Mozilla Public License 2.0
322 stars 85 forks source link

sort-ops tasks #115

Open jangorecki opened 4 years ago

jangorecki commented 4 years ago

As suggested by Matt, we should also present in an easy to grasp way what could be benefits on sorting data before doing various operations. In groupby task we do show timing of sorted data, but it is unknown how much time sorting takes. This issue is about adding new solution that simply sorts data as the first question and then proceeds with queries. In such a way it is easy what is the cost of sorting and benefits.