arrow-cls: work on benchmarks

[ivotron]

create a benchmark based on this https://blog.thedataincubator.com/2018/05/sqlite-vs-pandas-performance-benchmarks/ . Initially, we'll create a partitioning of parquet/ipc files and use that to benchmark IPC vs Parquet implementations. We'll then run this on rados (once it's ready)

[JayjeetAtGithub]

The IPC vs Parquet benchmarks can be found here.