AWS for DevOps — Lambda Basics

9.1 — Lambda Basics

AWS Lambda runs your code without provisioning or managing servers. Upload your code, configure a trigger, and Lambda handles everything — scaling, high availability, and patching.

How Lambda Works

  Event (API call, S3 upload, schedule)
       │
       ▼
  Lambda Service
  ┌──────────────────────────┐
  │  Provisions a container  │
  │  Loads your code         │
  │  Runs the handler        │
  │  Returns response        │
  │  Container stays warm    │ ◀── reused for next invocation
  └──────────────────────────┘
       │
       ▼
  Result (response, logs to CloudWatch)

Creating a Lambda Function

handler.py:

import json

def lambda_handler(event, context):
    """
    event  — the input data (JSON)
    context — runtime info (function name, memory, timeout remaining)
    """
    name = event.get('name', 'World')

    return {
        'statusCode': 200,
        'body': json.dumps({
            'message': f'Hello, {name}!'
        })
    }

# Zip the code
$ zip function.zip handler.py

# Create the function
$ aws lambda create-function \
    --function-name hello-world \
    --runtime python3.12 \
    --handler handler.lambda_handler \
    --role arn:aws:iam::123456789012:role/LambdaExecutionRole \
    --zip-file fileb://function.zip \
    --timeout 30 \
    --memory-size 256
{
    "FunctionName": "hello-world",
    "FunctionArn": "arn:aws:lambda:ap-south-1:123456789012:function:hello-world",
    "Runtime": "python3.12",
    "Handler": "handler.lambda_handler"
}

Invoking a Function

# Synchronous invocation
$ aws lambda invoke \
    --function-name hello-world \
    --payload '{"name": "DevOps Engineer"}' \
    --cli-binary-format raw-in-base64-out \
    output.json

$ cat output.json
{"statusCode": 200, "body": "{\"message\": \"Hello, DevOps Engineer!\"}"}

Lambda Configuration

Setting	Default	Max	Notes
Memory	128 MB	10,240 MB	CPU scales proportionally with memory
Timeout	3 seconds	900 seconds (15 min)	Function killed if exceeded
Ephemeral storage	512 MB	10,240 MB	/tmp directory
Concurrency	1,000	Account limit (request increase)	Concurrent executions
Package size	—	50 MB (zip), 250 MB (unzipped)	Use layers for large dependencies

Environment Variables

$ aws lambda update-function-configuration \
    --function-name hello-world \
    --environment 'Variables={DB_HOST=mydb.abc.rds.amazonaws.com,LOG_LEVEL=info}'

For secrets, use SSM Parameter Store or Secrets Manager with the Lambda Powertools library — never store secrets as plaintext environment variables.

Lambda Layers

Layers let you share code and dependencies across multiple functions:

# Create a layer from a zip (must follow directory structure)
# python/lib/python3.12/site-packages/<your packages>
$ zip -r my-layer.zip python/

$ aws lambda publish-layer-version \
    --layer-name my-shared-libs \
    --compatible-runtimes python3.12 \
    --zip-file fileb://my-layer.zip

# Attach layer to a function
$ aws lambda update-function-configuration \
    --function-name hello-world \
    --layers arn:aws:lambda:ap-south-1:123456789012:layer:my-shared-libs:1

Versions and Aliases

# Publish a version (immutable snapshot)
$ aws lambda publish-version --function-name hello-world
{
    "Version": "1",
    "FunctionArn": "arn:aws:lambda:ap-south-1:123456789012:function:hello-world:1"
}

# Create an alias pointing to version 1
$ aws lambda create-alias \
    --function-name hello-world \
    --name production \
    --function-version 1

# Shift traffic: 90% to v1, 10% to v2 (canary)
$ aws lambda update-alias \
    --function-name hello-world \
    --name production \
    --function-version 2 \
    --routing-config AdditionalVersionWeights={"1"=0.9}

Cold Starts

Aspect	Detail
What	First invocation spins up a new container (~100ms to 2s)
When	After idle period, or when scaling out new instances
Mitigation	Provisioned Concurrency, smaller packages, choose fast runtimes

# Provisioned Concurrency — keep N containers warm
$ aws lambda put-provisioned-concurrency-config \
    --function-name hello-world \
    --qualifier production \
    --provisioned-concurrent-executions 5

Common Mistakes

Mistake	Why It Hurts	Fix
Timeout too short	Function killed mid-execution	Set timeout based on actual execution time + buffer
Too little memory	Slow execution (CPU scales with memory)	Increase memory — often cheaper overall
Huge deployment package	Slow cold starts	Use layers, exclude dev dependencies
No dead letter queue	Failed async invocations lost	Configure DLQ (SQS or SNS)

Quick Reference

Task	Command
Create function	aws lambda create-function --function-name NAME --runtime python3.12 --handler handler.lambda_handler --role ROLE --zip-file fileb://code.zip
Invoke	aws lambda invoke --function-name NAME --payload '{}' output.json
Update code	aws lambda update-function-code --function-name NAME --zip-file fileb://code.zip
Update config	aws lambda update-function-configuration --function-name NAME --memory-size 512
View logs	aws logs tail /aws/lambda/NAME --follow

---

Next → API Gateway