Vectorized Query Execution in Apache Spark at Facebook Chen Yang Facebook

A standard query execution system processes one row at a time. Vectorized query execution batches multiples rows together in a columnar format, and each operator uses simple loops to iterate over data within a batch. This feature greatly reduces the CPU usage for reading, writing and query operations like scanning, filtering. In this talk, we will take a deep dive into Facebook's ORC-based vectorized reader and writer implementation, discuss how vectorization affects performance of various data types in Hive/Spark, and quantify the improvements vectorization brings to the Facebook Warehouse. About: Databricks provides a unified data analytics platform, powered by Apache Spark™, that accelerates innovation by unifying data science, engineering and business. Read more here: https://databricks.com/product/unified-data-analytics-platform Connect with us: Website: https://databricks.com Facebook: https://www.facebook.com/databricksinc Twitter: https://twitter.com/databricks LinkedIn: https://www.linkedin.com/company/databricks Instagram: https://www.instagram.com/databricksinc/ Databricks is proud to announce that Gartner has named us a Leader in both the 2021 Magic Quadrant for Cloud Database Management Systems and the 2021 Magic Quadrant for Data Science and Machine Learning Platforms. Download the reports here. https://databricks.com/databricks-named-leader-by-gartner