Parquet metadata as flatbuffers #9041
Description
Is your feature request related to a problem or challenge? Please describe what you are trying to do.
Reading wide tables spends relatively much time on decoding thrift metadata. There is an ongoing discussion and an active proposal in the Parquet community to address this by storing footer metadata as flatbuffers. See past mailing list discussions, parquet-format community discussion, active PR by @alkis and @Jiayi-Wang-db -> apache/arrow#48431 and Alkis' design doc.
Describe the solution you'd like
We would want to test an implementation Rust (and C++), to show performance and interoperability as well as have two reference implementations before discussing this further in the Parquet community.