Edit

Serverless Functions reference architectures

  • Article

A reference architecture is a template of required components and the technical requirements to implement them. A reference architecture isn't custom-built for a customer solution, but is a high-level scenario based on extensive experience. Before designing a serverless solution, use a reference architecture to visualize an ideal technical architecture, then blend and integrate it into your environment.

Common serverless architecture patterns

Common serverless architecture patterns include:

  • Serverless APIs, mobile and web backends.
  • Event and stream processing, Internet of Things (IoT) data processing, big data and machine learning pipelines.
  • Integration and enterprise service bus to connect line-of-business systems, publish and subscribe (Pub/Sub) to business events.
  • Automation and digital transformation and process automation.
  • Middleware, software-as-a-Service (SaaS) like Dynamics, and big data projects.

Web application backends the retail scenario: Pick up online orders from a queue, process them, and store the resulting data in a database.

Diagram shows a request made in a web app queued in Service Bus, which is then processed by a function and sent to Azure Cosmos DB.

Mobile application backends the financial services scenario: Colleagues use mobile banking to reimburse each other for lunch. Whoever paid for lunch requests payment through a mobile app, which triggers a notification on the colleagues' phones.

Diagram shows an H T T P A P I call, which is processed by a function and sent to Azure Cosmos DB. It triggers another function to send notifications.

IoT-connected backends in the manufacturing scenario: A manufacturing company uses IoT to monitor its machines. Functions detects anomalous data and that triggers a message to the service department when a repair is required.

Diagram shows I o T devices that produce requests for repair, which are sent to the I o T Hub, and are then routed for processing by using Zendesk.

Conversational bot processing for the hospitality scenario: Customers ask for available vacation accommodations on their phones. A serverless bot deciphers requests and returns vacation options.

Diagram shows a user request through a conversational interface that a bot deciphers for another function to process the request.

Real-time file processing for the healthcare scenario: The solution securely uploads patient records as PDF files. The solution then decomposes the data, by processes it using OCR detection, and it adds the data to a database for easy queries.

Diagram shows patient records uploaded, which is then decomposed and sent to Cognitive Services to be structured into a database.

Real-time stream processing for the independent software vendor (ISV) scenario: A massive cloud app collects huge amounts of telemetry data. The app processes that data in near real-time and stores it in a database for use in an analytics dashboard.

Diagram shows an app that collects data, which is ingested by Event Hubs, processed by a function, and sent to Azure Cosmos DB.

Scheduled task automation for the financial services scenario: The app analyzes a customer database for duplicate entries every 15 minutes, to avoid sending out multiple communications to the same customers.

Diagram shows a database that is cleaned by a function every 15 minutes, which removes duplicate entries.

Extending SaaS applications in the professional services scenario: A SaaS solution provides extensibility through webhooks, which Azure Functions can implement to automate certain workflows.

Diagram shows an issue created in GitHub, which triggers a webhook call. It's processed by a function that posts the issue details to Slack.

The following featured serverless reference architectures walk through specific scenarios. See the linked articles for architectural diagrams and details.

Serverless microservices

The serverless microservices reference architecture walks you through designing, developing, and delivering the Rideshare application by Relecloud, a fictitious company. You get hands-on instructions for configuring and deploying all the architectural components, with helpful information about each component.

Serverless web application and event processing with Azure Functions

This two-part solution describes a hypothetical drone delivery system. Drones send in-flight status to the cloud, which stores these messages for later use. A web application allows users to retrieve the messages to get the latest device status.

Event-based cloud automation

Automating workflows and repetitive tasks on the cloud can dramatically improve a DevOps team's productivity. A serverless model is best suited for event-driven automation scenarios. This event-based automation reference architecture illustrates two cloud automation scenarios: cost center tagging and throttling response.

Multicloud with Serverless Framework

The Serverless Framework architecture describes how the Microsoft Commercial Software Engineering (CSE) team partnered with a global retailer to deploy a highly-available serverless solution across both Azure and Amazon Web Services (AWS) cloud platforms, using the Serverless Framework.

More serverless Functions reference architectures

The following sections list other serverless and Azure Functions-related reference architectures and scenarios.

General

Web and mobile backend

AI + Machine Learning

Data and analytics

IoT

Gaming

Automation

Contributors

This article is maintained by Microsoft. It was originally written by the following contributors.

Principal author: