Explore multi-regional active-active resiliency in cloud architecture through this 30-minute conference talk from Strange Loop. Delve into the evolution from mainframe updates to microservices and cloud deployments, emphasizing the importance of preparing for region-wide failures. Examine considerations, tradeoffs, and challenges in building regionally-localized environments, including asynchronous data issues. Discover selected architecture patterns, new tools, and services to enhance service availability. Learn about calculating availability in series and parallel, various active architecture types, data replication strategies, tenant considerations, and essential patterns like Circuit Breaker and Exponential Backoff. Gain insights into worker queues, streams, and managed database solutions for easy data replication.
Overview
Syllabus
Intro
Calculating Availability (Series)
Calculating Availability (Paralel)
Basic Active Architecture
Basic Active-Passive Architecture
Basic Active Active Architecture N2
Data Replication (Active-Active Data Store)
Data Replication (Active-Passive Data Store)
Easy Data Replication Databases (Managed)
Considering Tenants - What are we failing over?
Circuit Breaker Pattern
Exponential Backoff Pattern
Worker Queues & Streams
Taught by
Strange Loop Conference
