Uber's Use of Presto for Data Analytics

This technical case study examines Uber's implementation of the Presto distributed query engine to enhance its data analytics capabilities. It outlines how Uber transitioned from traditional analytical databases to a more scalable solution that accommodates the vast amounts of data generated daily. The document details Uber's analytical model, which processes over 35 petabytes of data each day, and highlights the challenges faced in managing complex queries and ensuring security. It describes the evolution of Presto at Uber, including the automation of cluster management and workload management strategies to optimize performance. Additionally, the case study discusses the technical value of Presto, emphasizing its support for a wide range of analytical use cases and the benefits of open-source software development. Uber's contributions to the Presto project are also noted, showcasing their commitment to enhancing the platform for broader use. Overall, the case study provides a comprehensive look at how Uber leverages Presto to maintain its competitive edge in the analytics domain.