CapacityProviderConfig - Amazon Serverless Application Model
SyntaxPropertiesExamples
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

CapacityProviderConfig

Configures the capacity provider to which published versions of the function will be attached. This enables the function to run on customer-owned EC2 instances managed by Lambda.

Note

This configuration determines the compute type of a function and needs to be specified during the first function deployment. It can't be added or removed after the function resource has been created.

Syntax

To declare this entity in your Amazon Serverless Application Model (Amazon SAM) template, use the following syntax.

Properties

Arn

The ARN of the capacity provider to use for this function.

Type: String

Required: Yes

Amazon CloudFormation compatibility: This property is unique to SAM.

ExecutionEnvironmentMemoryGiBPerVCpu

The ratio of memory (in GiB) to vCPU for each execution environment.

Note

The memory ratio per CPU can't exceed function's total memory of 2048MB. The supported memory-to-CPU ratios are 2GB, 4GB, or 8GB per CPU.

Type: Float

Required: No

Amazon CloudFormation compatibility: This property is passed directly to the ExecutionEnvironmentMemoryGiBPerVCpu property of an AWS::Lambda::Function resource.

PerExecutionEnvironmentMaxConcurrency

The maximum number of concurrent executions per execution environment (sandbox).

Type: Integer

Required: No

Amazon CloudFormation compatibility: This property is passed directly to the PerExecutionEnvironmentMaxConcurrency property of an AWS::Lambda::Function resource.

Examples

Capacity provider configuration

The following example shows a capacity provider configuration that references a capacity provider resource.

CapacityProviderConfig:
  Arn: !GetAtt MyCapacityProvider.Arn
  ExecutionEnvironmentMemoryGiBPerVCpu: 4.0
  PerExecutionEnvironmentMaxConcurrency: 100