YouTube’s Procella Database — a new query processing engine built on top of various primitives proprietary to Google. This paper stood out to me for a number of reasons: this is the SQL query engine that runs YouTube, and it has been for a number of years now, and it contains optimizations that run queries faster than BigQuery, by two orders of magnitude in some cases. Every day this software serves 100s of billions of queries that span 10 PB of data. In this blog post, I’ll pick apart the paper and compare its descriptions to my understanding of various Hadoop-related technologies.
Learn faster. Dig deeper. See farther.
Join the O'Reilly online learning platform. Get a free trial today and find answers on the fly, or master something new and useful.
Observations on Technology Use in Hong Kong Protests (Maciej Ceglowski) — Telegram is the preferred messenger app among protesters. It’s used for one-on-one messaging between people, among small groups of people to coordinate, and among very large groups to amplify and disseminate information. The polls feature in Telegram is also a way of affirming consensus in group decision-making.
Misusing Personas with the Seven Dwarfs — “Dwarf personas” focus on users’ mental states and should help us understand how they might be personally impacted by the product. They appreciate that users are complex and can’t always be represented by a single persona.