Cloud Design Patterns

Empower security with the gatekeeper pattern by placing a layer in front of application services via API management, hiding backend endpoints, and enabling authentication with the application gateway.

Consolidate multiple service calls into a single gateway endpoint under the gateway aggregation pattern and return an aggregated response. Use fan-out/fan-in with Azure durable functions and emphasize latency, resiliency, tracing.

Offload authentication, authorization, token validation, encryption, ssl offloading, throttling, logging, and monitoring via the gateway offloading pattern to Azure Active Directory, Key Vault, Application Insights, Log Analytics, and API Management.

Learn the gateway routing pattern, using a single endpoint to route requests to multiple services via routing rules, with front door or application gateway and optional gateway aggregator.

Explore the priority queue pattern in cloud design patterns, applying high, medium, and low priorities to messages, using priority cues or dedicated queues, with implementation strategies and tradeoffs.

Learn how the publisher-subscriber pattern enables asynchronous messaging to multiple subscribers via an Azure Service Bus topic, with subscriptions, security, and handling duplicates and poisoned messages.

Explore the queue-based load leveling pattern that buffers database operations with a message queue between service and SQL database, boosting resiliency and performance while highlighting use cases and pitfalls.

Learn the asynchronous request-reply pattern to decouple front end from back end using azure service bus, enabling long-running processes and status checks with 202 accepted and 200 ok responses.

Isolate application components into bulkhead pools to prevent cascading failures and maintain service availability. Explore how bulkhead patterns pair with circuit breakers, throttling, and asynchronous communication using containers and workloads.

The retry pattern improves reliability by handling transient errors with server-side retries, using delay or immediate strategies and circuit breakers to avoid duplicates and user-visible failures.

Offload static content from compute services to cloud storage such as Azure Blob Storage or AWS S3, and deliver via a CDN to improve cost and performance.

Split large messages using the claim-check pattern by separating payload and check, store payload externally, and send a reference to enable scalable, reliable asynchronous processing.

Explore the ambassador pattern to centralize connectivity features in a dedicated container, simplifying remote service calls for microservices while highlighting latency and single points of failure.

Discover how the anti-corruption layer pattern enables safe integration between legacy and new systems by translating between protocols with an API, facade, and adapter, preventing corruption to each system.

The Strangler Fig pattern enables migrating a legacy system to a new architecture while keeping the website functional, via a facade and anti-corruption layer that isolates frontend from backend services.

Explore the backends-for-frontends pattern by splitting a single backend into desktop, mobile, and Alexa services to tailor data for each front end while centralizing business logic and reducing duplication.

Explore the sidecar pattern in cloud design: isolate modules in a separate container while enabling interaction with the primary application, improving resilience, configuration, and deployment across languages.

Explore the throttling pattern in cloud design, learn how to limit the rate of user requests to improve reliability and performance and protect service availability.

Learn the valet key pattern, a cloud design pattern that issues time-limited tokens to grant clients restricted access to cloud storage resources, offloading data movement from your app.

Federated identity delegates authentication to external providers, enabling single sign-on across apps, decoupling authentication from authorization, and simplifying user management while highlighting availability and security considerations.

Learn how deployment stamp pattern in Azure enables scalable, multiregion deployments with isolated stamps, using ARM templates, traffic routing via Azure Front Door, and considerations for cost and multitenancy.

Explore the geode pattern for global, geo-distributed architectures that scale out across regions with equal nodes, using Cosmos DB replication and Azure Front Door to route traffic for high availability.

Move external configuration data from local deployment settings to a cloud-based store to share across multiple apps and instances, improving manageability and avoiding per-instance reconfiguration.

The choreography pattern enables each service to participate in a business transaction using a message queue, removing a single point of control and promoting decentralized coordination.

Learn how the competing consumer pattern enables multiple consumer instances to pull from a single messaging channel, improving throughput, availability, and scalability while decoupling producers from consumers.

Learn how the cache-aside pattern loads on-demand data from the data store into the cache, with read-through, write-through, and write-behind strategies, eviction policies, and consistency considerations.

Explore the sequential convoy pattern for ordered message processing, using first-in, first-out queues to route categorized messages to dedicated consumers.

The compensating transaction pattern undoes work from a series of steps in an eventually consistent cloud workflow, using reverse operations to revert successful tasks when a failure occurs.