WSO2 Integrator on AWS

WSO2 Integrator provides support for building, deploying, securing, and observing integrations on Amazon Web Services (AWS). You can connect to AWS services, use Amazon Bedrock for AI capabilities, deploy integrations on AWS compute services, securely access AWS resources using IAM roles, and monitor integrations using AWS observability services. This page provides an overview of the capabilities and resources available for using WSO2 Integrator with AWS.

New to AWS?

Here are four AWS concepts you'll see throughout this guide

ECS

A fully managed container orchestration service for running your integration, used by the recommended deployment below.

IAM role

An identity AWS gives your integration, so it can use AWS without access keys in your code.

VPC

An isolated virtual network within AWS where you run and connect your resources.

Amazon Bedrock

The AWS service for using AI models from several providers through one API.

Start here

Four steps to unlock AWS Services with WSO2 Integrator

You need an AWS account and WSO2 Integrator installed.

1

Build

Connect to AWS services and add AI with Amazon Bedrock.

2

Deploy

Run it on Amazon ECS with Fargate, with no servers to manage.

3

Secure

Use IAM roles, not keys, and keep secrets out of code.

4

Observe

Send logs, traces and metrics to the AWS tools you already use.

Build for AWS

Decide how your integration interacts with AWS, add AI with Amazon Bedrock, and use Bedrock for AI assisted development.

Connect with AWS

Most teams start with a connector.

Connect to AWS services

Add a connector from the palette and sign in with the IAM role of wherever it runs.

Browse AWS connectors

React to AWS events

Start an integration when a message lands in an SQS queue, or when an event invokes an AWS Lambda function.

SQS listenerLambda event handlers

Call any AWS API

No connector for a service? Call its API directly - request signing (SigV4), Region and FIPS endpoints are handled for you.

Call AWS APIs

Build AI applications with Amazon Bedrock

Use AI models hosted on AWS, so your data stays in your AWS account.

Use Bedrock models

Call chat models and create embeddings through Amazon Bedrock, configured as a model or embedding provider.

Model providersEmbedding providersAuthentication

Answer from your own data (RAG)

Let AI answer from your documents using retrieval-augmented generation, with your source documents in S3.

Bedrock knowledge basesS3 VectorsS3 data loader

Run agents on Bedrock AgentCore

Deploy AI agents to AgentCore and give them tools through AgentCore Gateway.

DeployAuthenticationTracing

Give agents a memory

Store conversation memory in DynamoDB so agents remember earlier messages.

DynamoDB

Use Amazon Bedrock for AI-assisted development

Bedrock also powers AI features built into WSO2 Integrator itself.

Copilot with Amazon Bedrock

Sign in to WSO2 Integrator Copilot with your own Amazon Bedrock credentials.

Sign in with BedrockHow Copilot uses Bedrock

AI-assisted migration

Use Bedrock models to help migrate existing integrations onto WSO2 Integrator.

From MuleSoftFrom TIBCO

Deploy to AWS

Choose where your integration runs. Not sure? Start with ECS on Fargate.

Amazon ECS on Fargate Recommended

Run containers without managing any servers - the easiest place to start.

Amazon EKS

Choose this if your team already runs Kubernetes.

Amazon EC2

Choose this if you want a virtual machine you fully control.

AWS Lambda

Choose this for short, event-driven jobs that run only when needed.

Secure on AWS

Follow these three practices from day one.

Use IAM roles, not keys

Your integration picks up the role of wherever it runs - nothing secret lives in code.

Default credential chainAttach an IAM roleCross-account access

Keep secrets out of code

Load passwords and settings from AWS Secrets Manager or SSM Parameter Store.

Keep traffic private

Reach AWS services through VPC endpoints, with FIPS and dual-stack options.

Observe in AWS

Send logs, traces and metrics to the AWS tools you already use.

Logs

Collect logs in Amazon CloudWatch Logs.

CloudWatch Logs

Traces

Follow each request end to end with AWS X-Ray.

AWS X-Ray

Metrics

Track performance in Amazon Managed Service for Prometheus.

Managed Prometheus

AWS connectors

Each AWS service has its own connector in the palette. On AWS, set auth: DEFAULT_CREDENTIALS on a connector so it uses the IAM role of wherever it runs. See Access AWS services securely.

Category	Connector	Use it for
Messaging	Amazon SQS	Queue messages
	Amazon SNS	Publish to topics
	Amazon SES	Send email
Storage	Amazon S3	Bucket and object management
Database	Amazon DynamoDB	Read and write NoSQL data
	DynamoDB Streams	Follow table changes
	Amazon Redshift	Query data warehouses
	Redshift Data API	Query data warehouses over an API
	Amazon SimpleDB	Legacy key-value data
Security	AWS Secrets Manager	Read secrets at runtime
Marketplace	Metering	Meter usage for Marketplace products
	Entitlement	Check entitlements for Marketplace products
General-purpose	MySQL	For Amazon RDS and Aurora (MySQL-compatible)
	PostgreSQL	For Amazon RDS and Aurora (PostgreSQL-compatible)
	Redis	For ElastiCache and MemoryDB
Change data capture	CDC	Stream database changes, with schema history kept in S3

What's next?

Browse AWS connectors →

See every AWS connector available in WSO2 Integrator, from S3 and DynamoDB to SNS and SES.