Apache Arrow’s Final Frontier: Replacing Outdated Database Drivers

In the past decade, there has been a Cambrian-sized explosion of new analytics-oriented databases and query engines, yet the protocols these systems use to transfer data among one another were built for mining row-oriented transfers from transactional systems.

It’s a mismatch that slows down transfers and consumes excessive CPU usage.

Many of these analytical systems use ODBC (Open Database Connectivity), or its Java offshoot JDBC, as a conveyor, or DB-API for Python. But these all protocols copy data from the source by rows, and not by…