h2oai / db-benchmark

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

benchplot handle 1e9 k=2 q6-q10 #100

Closed jangorecki closed 5 years ago

jangorecki commented 5 years ago

Currently only spark and pydatatable are able to finish this complex data set, benchplot is not produced for that due to some exceptions handling. It probably could be produced.

Benchplot skipped as there are some questions not answered by any solutions for groupby G1_1e9_2e0_0_0

It would probably start working itself when spark or pydatatable could be capable to answer q6. Both does not have yet median function implemented.

jangorecki commented 5 years ago

for reference: https://h2oai.github.io/db-benchmark/groupby/plots/G1_1e9_2e0_0_0.advanced.png