Want to see how everyone will build serverless applications on AWS in 2025? Want to use those patterns today to gain a competitive advantage?
Come learn how, with App 2025! Every Thursday we'll build a bit of our application live on Twitch, so you can see the patterns that will help you get ahead today.
In the first episode of App 2025, viewers learn a high-level pattern for building serverless applications. A sample use case is presented: an order fulfillment application for AnyCompany , a fictional Software as a Service (SaaS) company. Amazon EventBridge is introduced as the backbone of the app. AWS Step Functions workflows, both standard and express, are introduced for business processes that consume events from EventBridge. Service integrations are introduced for data storage, and Identity and Access Management (IAM) considerations are discussed. Monitoring and auditability of events on EventBridge are discussed. Finally, AWS SAM CLI is introduced, and a simple template is created with a single resource, a custom EventBridge event bus.
In the second episode of App 2025, viewers learn a second high-level pattern for building serverless applications. AWS Step Functions Express Workflows are introduced for building high-volume, short-duration workflows. Express and Standard Workflows are compared and a rubric is provided for when to select each. The workflow event loop pattern is introduced: an event on an EventBridge event bus satisfies a rule and invokes an Express Workflow as a target. Once the workflow completes, a “ProcessCompleted” event is placed back onto the bus for further processing.
In the third episode of App 2025, viewers learn how to pause execution of AWS Step Functions workflows using callbacks. AWS Step Functions Standard Workflows are introduced for building lower-volume, longer-running workflows. The .waitTaskToken callback pattern is introduced for tasks that involve human interaction, are longer-running, or are cost-sensitive. Supported services are introduced, and SendTaskSuccess, SendTaskFailure, and SendTaskHeartbeat are discussed.
In the fourth episode of App 2025, viewers learn how to use Amazon SNS topics and Amazon SQS queues to simplify the App 2025 architecture. The stability of SNS and SQS is demonstrated by a review of their long history. SNS topics and subscriber endpoints are introduced, as are SQS standard and FIFO queues. The "buffered fanout" pattern is presented to demonstrate how SNS, SQS, and EventBridge complement one another.
In the fifth episode of App 2025, viewers learn how to use service integrations to store data persistently without invoking AWS Lambda functions. Amazon Kinesis Data Firehose is introduced as a serverless solution for ingesting streaming data and target for Amazon EventBridge rules. EventBridge input transformers are compared to Lambda functions as a mechanism for modifying data with Kinesis Data Firehose. The AWS Step Functions service integration with Amazon DynamoDB is discussed. Pass States are compared to Lambda functions as a mechanism for modifying data with Step Functions. An AWS SAM template is built that demonstrates how streaming data can be captured and stored for OLAP in Amazon S3 by Kinesis Data Firehose while concurrently being used for OLTP with Step Functions and DynamoDB – all serverlessly.
In the sixth episode of App 2025, viewers learn how to analyze and report the data generated by their serverless applications. Amazon Athena is introduced as a serverless solution for querying relational data stored in Amazon S3. The AWS Management Console is used to demonstrate how to access the data generated by Amazon Kinesis Data Firehose in the previous episode. Basic queries are written, including creating views. Amazon QuickSight is introduced for visualizing the data stored in S3, including creating data sets and visualizations. The Registry of Open Data on AWS is explored to enable viewers to augment and enhance their applications with publicly available data sets hosted on AWS. Finally, the material covered in the session and the series is reviewed.
- First, deploy the infrastructure.
- Next, deploy the customer microservice.
- Next, deploy the operations microservice.
- Next, deploy the billing microservice.
- Finally, deploy the simulator app.
Machine-generated transcripts for each episode are available in the transcripts directory.
Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. SPDX-License-Identifier: MIT-0