Apache Arrow Flight support

[SemanticBeeng]

Consider support for Apache Arrow Flight - data as a service paradigm

When combining "big data" / data science applications with microservices there is an impedance in how such big data is to be marshaled (both format and efficiency).

Apache Arrow aims to create a "shared runtime" between JVM, Python and R and used advanced techniques like "zero copy", memory mapping, etc.

This will allow writing distributed, polyglot algorithms (the "business logic" of data science apps)
https://www.slideshare.net/wesm/data-science-without-borders-jupytercon-2017
https://www.youtube.com/watch?v=wdmf1msbtVs&feature=youtu.be&t=790

Apache Arrow Flight aims to extend that performance to services.
https://www.slideshare.net/JacquesNadeau5/apache-arrow-flight-overview

Seems to me that Mu is a great place to introduce that support to the Scala world.

Please advise if this makes sense.