05. AWS Lambda Overview

AWS Lambda is a serverless compute service that lets you run code without provisioning or managing servers. It executes your code in response to events such as API calls, file uploads, or scheduled jobs.

---

🧩 EC2 vs Lambda

Feature	EC2	AWS Lambda
Type	Virtual servers in the cloud	Virtual functions – no servers to manage
Execution	Continuously running	Run on-demand
Limits	Bounded by RAM & CPU	Bounded by execution time
Scaling	Manual / Auto Scaling setup	Fully automated scaling
Billing	Pay for uptime (even if idle)	Pay only per request & duration

---

💡 Why AWS Lambda?

• 🧱 No Server Management – AWS handles provisioning and scaling.
• ⚙️ Automatic Scaling – Scales instantly from 1 request/day to thousands/second.
• 🔁 Event-Driven – Runs only when triggered by an event (e.g., S3 upload, API call).
• 💰 Simple Pricing – Pay per request and per compute time.
• 🧰 Fully Integrated – Works with other AWS services like S3, DynamoDB, API Gateway, CloudWatch.
• 📊 Easy Monitoring – Integrated with Amazon CloudWatch.
• 🚀 Flexible Resources – Up to 10GB RAM per function, with proportional CPU & network improvements.

---

🧠 Example Use Cases

🖼️ Serverless Thumbnail Creation

Step	Service	Action
1	Amazon S3	User uploads an image
2	AWS Lambda	Triggered by S3 upload event
3	Lambda Function	Generates thumbnail & metadata
4	Amazon S3 / DynamoDB	Stores thumbnail or metadata

✅ Fully serverless – no servers for S3, Lambda, or DynamoDB.

---

⏰ Serverless CRON Jobs

Use CloudWatch Events / EventBridge to trigger a Lambda function on a schedule (e.g., every hour or every day).
This replaces traditional Linux CRON jobs — completely serverless and automated.

---

🧑‍💻 AWS Lambda Language Support

Supported Languages:

• Node.js, Python, Java, C# (.NET Core / PowerShell), Ruby, Go,
• Custom Runtime (via API – e.g., Rust, Golang)
• Container Image (with Lambda Runtime API)

🟢 For exam purposes: Use ECS or Fargate for running general Docker containers, even though Lambda supports them.

---

💵 AWS Lambda Pricing

AWS Lambda charges you for how often your function runs (requests) and how long it runs (duration).
You get a free tier every month, and billing is extremely precise — per millisecond.

---

⚙️ Pricing Breakdown

Pricing Metric	Details
Requests	1M requests/month free, then $0.20 per 1M requests
Duration (GB-seconds)	400,000 GB-seconds/month free
Example	400,000 sec @ 1GB RAM OR 3,200,000 sec @ 128MB RAM
After Free Tier	$1 per 600,000 GB-seconds
Billing Granularity	Per 1 millisecond

---

🧠 How It Works

• Requests: Every time your function runs = 1 request.
• Duration: Based on execution time × memory size (GB-seconds).
• GB-second:
  • 1 second at 1GB = 1 GB-second
  • 1 second at 128MB = 0.125 GB-seconds

---

💰 Example

If your function runs 2 million times, uses 128 MB memory, and each run takes 1 second:

• Requests: 1M free → 1M paid = $0.20
• Duration: 2M × 1 sec × 0.125 GB = 250,000 GB-sec → within free tier

✅ Total Cost = $0.20

---

💡 Lambda pricing = pay per call + pay per duration
It’s highly cost-effective for event-driven, short, and scalable workloads.

---

🧾 Summary

• Serverless compute platform for running code in response to events.
• No servers, auto-scaling, event-driven, and cost-efficient.
• Pay only for what you use, with a generous free tier.
• Ideal for: automation scripts, backend APIs, data processing, image processing, and scheduled tasks.

---

5.1. Creating an AWS Lambda Function

This guide explains step-by-step how to create, configure, test, and monitor an AWS Lambda function using the AWS Management Console.

---

🧱 Step 1: Open the Lambda Console

1. Sign in to the AWS Management Console.
2. Search for Lambda in the search bar and open the AWS Lambda service.
3. Click Create function.

---

⚙️ Step 2: Choose a Creation Method

Select Author from scratch (you can also use a Blueprint, Container image, or upload existing code later).

• Function name: HelloLambda
• Runtime: Python 3.x (or Node.js, Java, etc.)
• Architecture: x86_64 (default)
• Permissions: Create a new role with basic Lambda permissions

---

🔐 Step 3: Set Permissions

• Lambda requires an IAM role to execute.
• Choose Create a new role with basic Lambda permissions.
  • This role allows your Lambda to write logs to CloudWatch.

You can later modify this role to grant access to:

• Amazon S3
• DynamoDB
• SNS / SQS
• Other AWS services

---

🧑‍💻 Step 4: Add Your Function Code

AWS automatically creates a sample Hello World function. Example (Python):

def lambda_handler(event, context):
    # Get the 'name' value from the event input
    name = event.get('name', 'Guest')

    # Log a message (goes to CloudWatch)
    print(f"Hello {name}, this is a simple Lambda test!")

    # Return a greeting message
    return {
        "message": f"Hello {name}, welcome to AWS Lambda!"
    }

Key points:

• Handler: lambda_handler — main entry point.
• Event: carries input data in JSON format.
• Context: contains runtime info (timeout, memory, etc.).

---

💾 Step 5: Deploy the Function

After editing or verifying the code:

• Click Deploy to save your Lambda code.
• Your function is now ready to invoke manually or via event triggers.

---

🧪 Step 6: Test the Function

1. Click Test (top-right).
2. Choose Configure test event.
3. Use sample event data:

{
  "name": "Himanshu"
}

4. Name the event (e.g., Test).
5. Click Save and Test.

Expected output: Hello Himanshu, welcome to AWS Lambda!

---

🧠 Step 7: Understand Function Results

• Execution results appear in the console's Execution results panel.
• Context and logs available for debugging.

Common metrics to check:

• Invocation time
• Memory usage
• Execution duration

---

📊 Step 8: Monitor Function Activity

Go to the Monitor tab to view:

• Invocation count
• Duration
• Errors
• Throttles

Click View logs in CloudWatch to see detailed logs like:

---

⚙️ Step 9: Configure Function Settings

You can customize:

• Memory (MB): allocate between 128 MB – 10 GB
• Timeout: max execution time before Lambda stops
• Ephemeral storage: temporary space (default 512 MB, max 10 GB)
• Environment variables: define configuration parameters
• VPC: connect Lambda to private networks

---

🔗 Step 10: Add Event Triggers

Connect Lambda to various event sources:

• Amazon S3 (on file uploads)
• DynamoDB Streams
• API Gateway
• EventBridge
• SNS / SQS
• Kinesis Data Streams

Example flow: S3 → PutObject → triggers Lambda → Lambda processes file/image

---

🔐 Step 11: IAM Role and Permissions

1. Go to Configuration → Permissions.
2. Click the role name to open it in the IAM Console.
3. Review attached policy, e.g., AWSLambdaBasicExecutionRole — grants CloudWatch logging permissions.
4. Add more permissions as required (e.g., S3 access).

---

✅ Step 12: Summary

You have successfully:

• Created a Lambda function (HelloWorld)
• Deployed and tested it
• Viewed CloudWatch logs and metrics
• Learned how to add triggers and manage permissions

AWS Lambda lets you run event-driven, auto-scaling, pay-per-use code without managing servers.