Micah Kornfield created ARROW-8491: -------------------------------------- Summary: [C++][Parquet] Add benchmarks for rep/def level decoding at multiple levels Key: ARROW-8491 URL: https://issues.apache.org/jira/browse/ARROW-8491 Project: Apache Arrow Issue Type: Sub-task Components: C++ Reporter: Micah Kornfield
using synthetic data would be a good start but starting to build a small corpus of 1-10 MB files in general to run against would be a good idea. -- This message was sent by Atlassian Jira (v8.3.4#803005)