Explore Conflict-free Replicated Data Types (CRDTs) and their application in building highly available decentralized services using Go in this GopherCon 2021 conference talk. Learn about the fundamentals of CRDTs, including state-based and operation-based approaches, and discover how Bread implemented them to achieve eventual consistency in distributed systems. Gain insights into the advantages and disadvantages of various CRDT types, such as counters and PM counters, and understand their practical implementation in Go. Delve into topics like contextual series, delivery mechanisms, integration testing, and best practices for leveraging CRDTs in your own distributed systems projects.
Using CRDTs to Build a Highly Available Decentralized Service
Overview
Syllabus
Introduction
Outline
Consistency
Conflictfree replicated data types
Statebased vs operationbased
Replicas
Counters
PM Counter
Advantages Disadvantages
Operation Based
Clock
Contextual Series
Conclusion
Delivery
Miscellaneous
Integration Testing
Conclusions
Taught by
Gopher Academy
Related Courses
-
CRDTs Illustrated
-
Living in Eventually Consistent Reality
-
Mixed and Verified Consistency with Propel and ConOpY
-
Conflict-Free Replicated Data Types (CRDTs) in Distributed Systems
-
Building High Available Systems with Akka Distributed Data
-
Easy Eventual Consistency with Actor Models + Amazon Web Services
