AWS Lambda Functions
Run code at any scale. No servers or clusters to manage. Pay-per-use billing.
Why Lambda Functions?
Focus on your code and let Lambda handle the infrastructure. Write a handler, connect it to any of 220+ AWS event sources, and Lambda runs it. No servers to provision, no scaling to configure. Lambda automatically scales to match demand, with pay-per-use billing down to the millisecond.
Run for up to 15 minutes per invocation, or orchestrate multi-step workflows lasting up to a year with Lambda durable functions. Choose on-demand execution for variable workloads, or Lambda Managed Instances for sustained, compute-intensive throughput.
What do you get with Lambda Functions?
Scale and integration
Open allLambda automatically scales to match the rate of incoming requests and scales down to zero during idle periods, requiring no manual configuration. Each function can scale up to 1,000 concurrent executions every 10 seconds incrementally, until reaching the account's concurrency limit.
Lambda natively integrates with over 220 AWS services and 50 SaaS applications, enabling faster application composition without the need to write custom integration code.
Lambda runs on highly available infrastructure across multiple Availability Zones. Strengthen application resilience with AWS Lambda durable functions, which provides built-in error handling, automatic retries, and recovery after failures.
Developer experience
Open allLambda natively supports Python, Node.js, Java, C#, Go, and Ruby, as well as custom runtimes. Package your code as ZIP files or container images, and deploy through the AWS Console, Infrastructure as Code, or CI/CD pipelines.
AWS Lambda integrates with a comprehensive set of security, observability, and compliance services, enabling alerts or automatic remediation of security threats. AWS Lambda provides built-in monitoring capabilities by automatically capturing and sending logs, metrics, and traces to AWS monitoring and observability services such as Amazon CloudWatch and AWS X-Ray. With Application Signals, AWS Lambda offers an out-of-the-box application performance monitoring (APM) tool to track throughput, availability, latency, faults, and errors in your serverless applications. Additionally, AWS Lambda supports seamless integration with third-party monitoring tools through Lambda extensions, making it easy to use your preferred tools for monitoring and troubleshooting your serverless applications.
Author, build, debug, test, and deploy functions directly from your IDE with dedicated toolkits for VS Code and other editors. Lambda supports remote debugging and local testing of cloud functions, enabling rapid iteration from your local environment.
AI coding agents can accelerate serverless development by providing intelligent assistance for application scaffolding, deployment configuration, debugging, and infrastructure-as-code generation. The AWS Serverless MCP Server equips agents like Claude Code, Kiro, Cursor, GitHub Copilot, Windsurf, and others with the tools and architectural guidance to build and manage Lambda applications. Pair it with the AWS serverless skill from the Agent Toolkit for AWS for a complete AI-assisted development workflow.
Performance
Open allKeep functions initialized and ready to respond, ensuring consistent start-up performance with minimal cold start latency.
Delivers up to 10x faster startup for Java, and from several seconds to sub-second for Python and .NET, with minimal or no changes to your function code.
Configure throughput for event source connectors, optimizing performance for Apache Kafka sources including Amazon MSK and self-managed Kafka.
Security and compliance
Open allLambda executes functions on dedicated compute resources powered by the AWS Nitro System, never shared between functions, customers, or accounts. With tenant isolation mode, Lambda isolates request processing for individual tenants, making it simple to build multi-tenant applications with execution environment isolation at the tenant level. Execution environments operate within network-isolated VPCs managed by Lambda, where network ingress is strictly limited to Lambda-owned services and accounts.
Lambda is certified under SOC, HIPAA, PCI, ISO, and other compliance programs. See the full services in scope for the latest certifications.
Cost efficiency
Open allYou are charged only for the compute time your code consumes. No minimum fees, no upfront commitments.
Run on x86 or Arm-based (Graviton) processors. Switch between architectures without code changes to optimize for cost or performance.
Lambda offers tiered pricing for on-demand duration above certain monthly thresholds. Compute Savings Plans offer up to 17% savings on duration and Provisioned Concurrency with a one- or three-year commitment.
Extend Lambda's operational simplicity to EC2 compute instances for specialized workloads. Lambda fully manages instance lifecycle, OS patching, routing, load balancing, and auto scaling. You keep the familiar Lambda programming model while benefiting from EC2's pricing advantages.