Speaker

Kay Ousterhout, PhD Student at UC Berkeley

Kay Ousterhout

PhD Student, UC Berkeley

Kay Ousterhout is a Spark committer and PMC member and a PhD student at UC Berkeley. In the Spark project, Kay is a committer of the scheduler, and her work on Spark has focused on improving scheduler performance. At UC Berkeley, Kay’s research focuses on understanding and improving performance of large-scale analytics frameworks.

Sessions

Re-Architecting Spark For Performance Understandability

This talk will describe Monotasks, a new architecture for the core of Spark that makes performance easier to reason about. In Spark today, pervasive parallelism and pipelining make it difficult to answer even simple performance… Read more