Cloud Native Patterns – Shifting Responsibilities
summary
Claim: moving cloud‑native patterns from the app tier into distributed SQL databases reduces developer burden by embedding redundancy, routing, and resilience in the data layer. Shows Kubernetes replica‑set reconciliation and health checks, Google Spanner’s replication and synchronous writes, Netflix’s 2015 outage caused by a retry storm as evidence that circuit breakers and retries matter, and that 100 microservices each at 99.999% availability compound into ~43 minutes/month downtime at the homepage. Argues sharding/routing, client/server load balancing and caches shift cleanly into the database.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 3:10Cascading Requests
- 6:20Multiple Instances and Load Balancing
- 12:28Retries
- 13:53Retry Storm
- 16:00Data Redundancy
- 17:57Statelessness
- 26:49Summary
related talks
28mCloud Native Tech: The Building Blocks of Modern DB Software
44mRunning Non-Cloud-Native Databases in Cloud-Native Environments
29mFollowing the Data: A Deeper Look into Distributed Applications
28mConnectivity Patterns for Cloud Native Applications
32mCloud-Native Data: What is it? - Majors, Dunlap, O’Grady, Richardson, Drewitz
26mCloud Native Database Meetup Japan: The State of Distributed Database in Japan
45mFrom Laptop to Cloud: Developing Cloud Native Applications with Containerized Databases
24mCloud Native Hacks for the Real World: Solve your daily struggles with tech