Eliminate Data Transfer Bottlenecks with Apache Arrow Flight

Session Abstract

Data scientists and analysts need fast access to millions of records. However, the throughput of traditional data transfer APIs such as pyodbc is insufficient to meet their needs, so they extract data into files on their local machines. This results in significant security and governance issues as well as stale data copies.

Join us as we explore how Apache Arrow Flight solves this problem by providing a new and modern standard for transporting large data between networked applications. We’ll even run a live bake-off to demonstrate how Arrow Flight enables more than 10x faster transfer rates for highly parallel systems compared to pyodbc.

Presented By

Wes McKinney

Apache Arrow PMC Chair and Co-Creator, Ursa Computing CEO and Founder

Matt Topol

Apache Arrow Contributor, Principal Software Engineer at FactSet

Kyle Porter

Apache Arrow Contributor, Dremio Architect

Need Some Help?

Contact us Ask the Community