AWS Lambda Deployment
Deploy GLACIS sidecars on AWS Lambda for event-driven, pay-per-use attestation infrastructure.
Why Lambda?
- Serverless: No infrastructure management
- Pay-per-use: Only pay for compute time used
- AWS integration: Native integration with API Gateway, secrets
- Global: Deploy to any AWS region
Prerequisites
- AWS CLI configured
- AWS SAM CLI installed
- GLACIS organization and API key
Quick Start
-
Clone the sidecar template
Terminal window git clone https://github.com/glacis-io/sidecar-lambda-templatecd sidecar-lambda-template -
Store secrets in Parameter Store
Terminal window aws ssm put-parameter --name /glacis/api-key --value "glc_your_api_key" --type SecureStringaws ssm put-parameter --name /glacis/openai-key --value "sk-your-key" --type SecureString -
Deploy with SAM
Terminal window sam buildsam deploy --guided -
Get API Gateway endpoint
Terminal window aws cloudformation describe-stacks --stack-name glacis-sidecar \--query 'Stacks[0].Outputs[?OutputKey==`ApiEndpoint`].OutputValue' --output text
SAM Template
AWSTemplateFormatVersion: '2010-09-09'Transform: AWS::Serverless-2016-10-31
Resources: GlacisSidecar: Type: AWS::Serverless::Function Properties: Handler: index.handler Runtime: nodejs20.x Timeout: 30 MemorySize: 512 Environment: Variables: GLACIS_ORG_ID: !Ref OrgId Events: Api: Type: HttpApiPerformance
| Metric | Value |
|---|---|
| Cold start | ~100ms |
| Request overhead | ~15ms |
| Memory | 512MB recommended |
| Timeout | 30s max |