Principles of the CAP Theorem and Distributed Data

Description

The CAP Theorem states that it is impossible for a distributed database to simultaneously guarantee two of the following: Consistency, Availability, or Partition Tolerance. The widely accepted theorem is often used to analyze and position new databases. For more details about CAP Theorem check out this blog from our Chief Architect and co-founder, Ben Darnell: https://www.cockroachlabs.com/blog/limits-of-the-cap-theorem/ In this session, we explore the three guarantees, delivering a definition for each, and outlining how each might affect your distributed system. This is a rather academic talk and, hopefully, it will help you understand the history and how to use this important concept. We cover: What are the challenges to ensuring consistent distributed data? What are the availability challenges for a distributed database? What are the challenges to dealing with network failures and dropped messages? RAFT, MVCC and other algorithms We will talk through CP and CA databases and how they are used. How CockroachDB delivers on these key traits.