Cloud deployment architectures define how and where cloud workloads are distributed across geographic regions, availability zones, and infrastructure types to meet requirements for availability, latency, compliance, and disaster recovery. These architectures range from simple single-region deployments to complex geo-distributed systems spanning multiple clouds and edge locations. In 2026, AI/ML workloads, WebAssembly runtimes, and eBPF-based networking have added new deployment dimensions, while GitOps, platform engineering, and FinOps practices have matured into essential operational standards. Understanding the trade-offs between cost, complexity, recovery objectives, and data residency is critical — each architecture pattern addresses different failure domains, consistency requirements, and user proximity needs.
What This Cheat Sheet Covers
This topic spans 16 focused tables and 120 indexed concepts, 114 flashcards. Below is a complete table-by-table outline of this topic, spanning foundational concepts through advanced details.
A jump-to index of every table row in this cheat sheet.
An interactive map of every table and concept in this topic.
Table 1: Regional Deployment Patterns
The first architectural decision is geographic: how widely you spread a workload across zones, regions, and continents. Each step up the ladder — from a single zone to multi-AZ to multi-region to fully geo-distributed — buys you more resilience and lower latency for distant users, but adds cost and the headache of keeping data in sync across the gap.
| Pattern | Example | Description | |
|---|---|---|---|
Resources across us-east-1a, us-east-1b, us-east-1c | • Distributes workloads across multiple isolated data centers within one region • protects against data center failures but not regional disasters • most common production baseline. | ||
Active in us-east-1, standby in eu-west-1 | • Workloads span two or more geographic regions • provides disaster recovery for regional outages and reduces latency for global users • increases cost and operational complexity. | ||
App nodes in North America, Europe, Asia | • Workloads deployed across multiple continents • minimizes global user latency via proximity routing • enables data residency compliance • requires cross-region data synchronization. | ||
App deployed in us-east-1 only | • All resources in one geographic region • simplest and lowest-cost but no regional disaster recovery • a regional outage causes total unavailability. | ||
AWS Local Zone in Los Angeles for low-latency gaming | • Cloud compute extended to metro or edge locations managed by a public cloud provider • single-digit ms latency to local users • ideal for media, gaming, and real-time applications. | ||
All VMs in one availability zone | • Resources constrained to one availability zone • lowest latency within zone but no resilience to zone failures • used for cost optimization in non-critical workloads. |