The AWS adapter takes a Spring Cloud Function app and converts it to a form that can run in AWS Lambda.


The adapter has a couple of generic request handlers that you can use. The most generic is SpringBootStreamHandler, which uses a Jackson ObjectMapper provided by Spring Boot to serialize and deserialize the objects in the function. There is also a SpringBootRequestHandler which you can extend, and provide the input and output types as type parameters (enabling AWS to inspect the class and do the JSON conversions itself).

If your app has more than one @Bean of type Function etc. then you can choose the one to use by configuring function.name (e.g. as FUNCTION_NAME environment variable in AWS). The functions are extracted from the Spring Cloud FunctionCatalog (searching first for Function then Consumer and finally Supplier).

Notes on JAR Layout

You don’t need the Spring Cloud Function Web or Stream adapter at runtime in Lambda, so you might need to exclude those before you create the JAR you send to AWS. A Lambda application has to be shaded, but a Spring Boot standalone application does not, so you can run the same app using 2 separate jars (as per the sample). The sample app creates 2 jar files, one with an aws classifier for deploying in Lambda, and one executable (thin) jar that includes spring-cloud-function-web at runtime. Spring Cloud Function will try and locate a "main class" for you from the JAR file manifest, using the Start-Class attribute (which will be added for you by the Spring Boot tooling if you use the starter parent). If there is no Start-Class in your manifest you can use an environment variable MAIN_CLASS when you deploy the function to AWS.


Build the sample under spring-cloud-function-samples/function-sample-aws and upload the -aws jar file to Lambda. The handler can be example.Handler or org.springframework.cloud.function.adapter.aws.SpringBootStreamHandler (FQN of the class, not a method reference, although Lambda does accept method references).

./mvnw -U clean package

Using the AWS command line tools it looks like this:

aws lambda create-function --function-name Uppercase --role arn:aws:iam::[USERID]:role/service-role/[ROLE] --zip-file fileb://function-sample-aws/target/function-sample-aws-2.0.0.BUILD-SNAPSHOT-aws.jar --handler org.springframework.cloud.function.adapter.aws.SpringBootStreamHandler --description "Spring Cloud Function Adapter Example" --runtime java8 --region us-east-1 --timeout 30 --memory-size 1024 --publish

The input type for the function in the AWS sample is a Foo with a single property called "value". So you would need this to test it:

  "value": "test"
The AWS sample app is written in the "functional" style (as an ApplicationContextInitializer). This is much faster on startup in Lambda than the traditional @Bean style, so if you don’t need @Beans (or @EnableAutoConfiguration) it’s a good choice. Warm starts are not affected.

Functional Bean Definitions

Your functions will start much quicker if you can use functional bean definitions instead of @Bean. To do this make your main class an ApplicationContextInitalizer<GenericApplicationContext> and use the registerBean() methods in GenericApplicationContext to create all the beans you need. You function need sto be registered as a bean of type FunctionRegistration so that the input and output types can be accessed by the framework. There is an example in github (the AWS sample is written in this style). It would look something like this:

public class FuncApplication implements ApplicationContextInitializer<GenericApplicationContext> {

	public static void main(String[] args) throws Exception {
		FunctionalSpringApplication.run(FuncApplication.class, args);

	public Function<Foo, Bar> function() {
		return value -> new Bar(value.uppercase()));

	public void initialize(GenericApplicationContext context) {
		context.registerBean("function", FunctionRegistration.class,
			() -> new FunctionRegistration<Function<Foo, Bar>>(function())


Platfom Specific Features

HTTP and API Gateway

AWS has some platform-specific data types, including batching of messages, which is much more efficient than processing each one individually. To make use of these types you can write a function that depends on those types. Or you can rely on Spring to extract the data from the AWS types and convert it to a Spring Message. To do this you tell AWS that the function is of a specific generic handler type (depending on the AWS service) and provide a bean of type Function<Message<S>,Message<T>>, where S and T are your business data types. If there is more than one bean of type Function you may also need to configure the Spring Boot property function.name to be the name of the target bean (e.g. use FUNCTION_NAME as an environment variable).

The supported AWS services and generic handler types are listed below:

Service AWS Types Generic Handler

API Gateway

APIGatewayProxyRequestEvent, APIGatewayProxyResponseEvent





For example, to deploy behind an API Gateway, use --handler org.springframework.cloud.function.adapter.aws.SpringBootApiGatewayRequestHandler in your AWS command line (in via the UI) and define a @Bean of type Function<Message<Foo>,Message<Bar>> where Foo and Bar are POJO types (the data will be marshalled and unmarshalled by AWS using Jackson).

Custom Runtime

An AWS Lambda custom runtime can be created really easily using the HTTP export features in Spring Cloud Function Web. To make this work just add Spring Cloud Function AWS and Spring Cloud Function Web as dependencies in your project and set the following in your application.properties:


Set the handler name in AWS to the name of your function. Then provide a bootstrap script in the root of your zip/jar that runs the Spring Boot application. The functional bean definition style works for custom runtimes too, and is faster than the @Bean style, so the example FuncApplication above would work. A custom runtime can start up much quicker even than a functional bean implementation of a Java lambda - it depends mostly on the number of classes you need to load at runtime. Spring doesn’t do very much here, so you can reduce the cold start time by only using primitive types in your function, for instance, and not doing any work in custom @PostConstruct initializers.