Why Databricks Migrated to a Scalable Data Architecture

Overview

Databricks was using MySQL to manage its control plane containing multiple cloud services supporting users, cluster management, and web applications. However, as the company’s cloud usage grew by 3x, MySQL was unable to handle the increased load: queries became slow or even unresponsive for large customers.

In this short video, Databricks Staff Software Engineer Min Zhou sheds light on how the company’s migration to TiDB database, an advanced open-source, distributed SQL database with lowered hardware costs while increasing QPS performance by 10x.