Build high-performance apps with AWS Lambda Managed Instances

This article explains how to build high-performance applications using AWS Lambda Managed Instances (LMI), which runs Lambda functions on selected EC2 instance types while maintaining the Lambda programming model.

• Create Capacity Providers defining infrastructure blueprint with instance types, VPC configuration, and scaling behavior
• Choose from 400+ EC2 instance types: C (compute), M (general purpose), R (memory optimized)
• Configure memory-to-vCPU ratios: 2:1 (compute), 4:1 (general), 8:1 (memory optimized)
• Support multiple concurrent invocations per execution environment to reduce costs
• Publish function versions to provision EC2 instances and pre-warm execution environments
• Lambda automatically scales based on CPU utilization up to configured MaxVCpuCount limit
• Code must be thread-safe for concurrent execution across shared environments
• Monitor CloudWatch metrics for CPU, memory, concurrency, and disk utilization
• Supports I/O-bound and CPU-intensive workloads with optimized configurations

LMI combines serverless simplicity with EC2 compute flexibility for data pipelines, compute-intensive operations, and cost-sensitive steady-state applications.