AWS Step Functions

Seamless integration of agentic workflows across public and private endpoints with human-in-the-loop controls and automated failure recovery

Keep application logic strictly separated from implementation, allowing changes to workflows without affecting business logic

HIPAA eligible service for applications containing healthcare-related information and personal health information

Compliant with SOC (System & Organization Control) measures with third-party audit results available

Built-in state management that maintains application state during execution and tracks workflow progress

Store and reference payload data easily within workflows using Variables, assign and reference data at any state without passing through intermediate states

Built-in data transformation capabilities including date/time formatting and mathematical operations to reduce custom code

Automatic try/catch and retry mechanisms for handling exceptions and failures across workflow steps

Built-in timeout capabilities for managing long-running tasks and preventing indefinite waits

Coordinate applications that can make HTTPS connections regardless of hosting location including EC2, mobile devices, and on-premises servers

Integration with over 10,000 APIs created by Amazon API Gateway

Direct integration with 220+ AWS services including Lambda, ECS, DynamoDB, SNS, SQS, Athena, Batch, Glue, EMR, SageMaker, and more

Integration with Amazon EventBridge to coordinate events from over a hundred AWS and third-party SaaS event sources

Seamless integration with SaaS applications and external APIs directly from workflows

Integration with AWS IoT Rules Engine actions for IoT workflow orchestration

Coordinate existing Lambda functions and microservices into robust applications and rewire them into new compositions

Integration with AWS CloudTrail for auditing and compliance tracking of workflow operations

Integration with Amazon CloudWatch for monitoring and logging of workflow executions

Real-time diagnostics with complete execution history including overall state, failed steps, inputs, and outputs

Console-based dashboard for monitoring workflow execution with visual step progression and error highlighting

Automatic scaling of operations and underlying compute in response to changing workloads

Advanced map state for distributed processing of large datasets with high concurrency and batching support

High-volume workflow execution supporting event rates greater than 100,000 per second for IoT and data ingestion

Large-scale parallel processing capability to iterate over objects like images, logs, or CSV files and launch thousands of concurrent workflows

Execute multiple paths in parallel to handle concurrent operations within workflows

Built-in fault tolerance with service capacity across multiple Availability Zones to protect against machine or data center failures

Integration with AWS Identity and Access Management for least-privileged access control

Integration with private networks and on-premises systems using AWS PrivateLink and Amazon VPC Lattice

Support for VPC Endpoints using AWS PrivateLink to access Step Functions from VPC-enabled services without traversing public internet

Define workflows as state machines with easy-to-understand visual diagrams and drag-and-drop interface

Support for manual approval workflows in security incident response and other processes requiring human review

Standard workflow type for long-running executions with state transitions pricing model