lambda on Alperen Doc

aws-lambda

Mon, 01 Jan 0001 00:00:00 +0000

AWS Lambda#

So you’ve got a simple application you want to run in the cloud. Maybe it’s an API endpoint, a scheduled task, or a function that processes files uploaded to Amazon S3. You could spin up an EC2 instance, set up an ECS cluster, or run code directly on a server.

AWS Lambda is a serverless compute service. You write your code and upload it, and AWS handles the rest: provisioning servers, scaling, patching, and monitoring. You only pay for the compute time you actually use.

deploy-lambda

Mon, 01 Jan 0001 00:00:00 +0000

Deploy Lambda#

AWS Lambda handlers in Python are just functions that:

Accept an event (input data)
Return a response object

In other words, we don’t worry about the runtime, the server, or even any sort of HTTP library. We just write a function that accepts an event object and returns a response object.

The process to create a new Lambda function is simple:

Create an IAM role that gives Lambda permission to write logs to CloudWatch
Create the Lambda function itself
Paste the handler code and deploy

Assignment#

Create and deploy patientping-ip-function with role patientping-lambda-role by pasting Python code in the AWS Console.

testing-lambda-function

Mon, 01 Jan 0001 00:00:00 +0000

Testing Lambda Functions#

Your Lambda function is deployed, now let’s test it. Lambda provides several ways to invoke functions, and we’ll explore a few.

When you test a Lambda function directly (via the CLI or AWS Console), you’re not making an HTTP request to it. You’re telling AWS to execute your code with a specific payload. Take a look at this diagram:

Invoke from the CLI#

For our function, we need to simulate what API Gateway would send. When you invoke the function directly via the console, you’re providing mock data. The IP address in your test event is not your real IP address; it’s example data you’re feeding to the function.

lambda-api-gateway

Mon, 01 Jan 0001 00:00:00 +0000

API Gateway#

Your Lambda function works, but it can only be invoked from the AWS CLI or AWS Console. To make it accessible over HTTP (so you can call it from a browser, mobile app, or any HTTP client), we need Amazon API Gateway.

What Is API Gateway?#

API Gateway is a managed service that creates, publishes, and manages APIs. It sits in front of your Lambda function and handles much of what a web server or proxy would do. Have a look at this diagram:

cloudwatch-log-groups-and-viewing-lambda-logs

Mon, 01 Jan 0001 00:00:00 +0000

CloudWatch Log Groups and Viewing Lambda Logs#

When something goes wrong with your Lambda, how do you see what’s happening inside the function? Debugging Lambda is trickier than a local app: you can’t just attach a debugger to a process. There are ways to run Lambda locally (e.g., SAM CLI) for breakpoints, or AWS Toolkit for VSCode for IDE integration. For day-to-day debugging, though, you’ll rely on logs.

Lambda automatically sends everything your function writes to stdout and stderr (including print() in Python) to Amazon CloudWatch Logs. As long as your Lambda has the basic execution role (which we attached in the deploy lesson), it can write logs. AWS creates a log group for your function and streams each run’s output there.

other-lambda-use-cases

Mon, 01 Jan 0001 00:00:00 +0000

Other Lambda Use Cases#

We built a simple HTTP API with Lambda, but that’s just scratching the surface. Lambda’s real power comes from its ability to automatically trigger handlers based on events from across AWS and beyond.

S3 Triggers: Automatic Image Processing#

One popular Lambda use case is processing files when they’re uploaded to Amazon S3. For example:

A user uploads an image to S3 bucket user-uploads
S3 sends an event to Lambda: “A new file was uploaded”
A Lambda function runs that:
- Downloads the image from S3
- Resizes it to thumbnail size (using an image processing library)
- Uploads the thumbnail to another S3 bucket user-thumbnails
Done!

No servers to maintain, and you only pay when images are actually uploaded. If you get 1,000 uploads one day and 10 the next, Lambda automatically scales up and down.