📸 Serverless Image Resizer (AWS Free Tier Project)

🚀 Project Overview

The Serverless Image Resizer is an event-driven cloud application that automatically resizes images when they are uploaded. It uses AWS serverless services to process and deliver optimized images without managing any servers.

When a user uploads an image, it is automatically resized and stored in a separate bucket. The processed image is then served globally using a CDN for fast access.

---

🎯 Key Features

• Automatic image resizing on upload
• Fully serverless architecture
• Scalable and cost-efficient (AWS Free Tier compatible)
• Global image delivery using CDN
• No manual processing required

---

🧱 Architecture

---

🛠️ Technologies Used

• Amazon S3 – Storage for input and output images
• AWS Lambda – Serverless compute for resizing images
• Pillow (Python Library) – Image processing
• Amazon CloudFront – Content Delivery Network (CDN)
• IAM Roles – Secure permissions management

---

⚙️ Workflow

1. User uploads an image to the S3 Input Bucket

2. S3 triggers a Lambda function

3. Lambda:

   • Downloads the image
   • Resizes it (e.g., 300x300)
   • Optimizes the format

4. Resized image is saved to S3 Output Bucket

5. CloudFront serves the image globally

---

⚙️ Setup Instructions

1. Create S3 Buckets

• Input bucket
• Output bucket

2. Create Lambda Function

• Runtime: Python 3.10
• Add IAM Role with S3 access

3. Add Pillow Layer

(CloudShell method)

4. Configure S3 Trigger

5. Setup CloudFront

• Origin: Output bucket
• Enable OAC

🧩 Why We Use a Lambda Layer (Pillow)

AWS Lambda provides a minimal runtime environment and does not include image processing libraries like Pillow by default.

Since this project requires:

• Image resizing
• Image format conversion
• Image optimization

👉 We must include an external library (Pillow).

✅ Why Lambda Layer?

• Keeps Lambda deployment package small
• Reusable across multiple functions
• Cleaner architecture
• Follows AWS best practices

---

⚠️ Important Note

Pillow contains compiled components (_imaging) which must match:

• Linux environment (Amazon Linux)
• Correct Python runtime (3.10 in this project)

👉 That’s why we use CloudShell to build the layer.

---

🚀 Create Pillow Layer Using CloudShell (Recommended)

Step 1: Open CloudShell

Go to AWS Console → Click CloudShell (>_)

---

Step 2: Run Commands

# Clean setup
rm -rf pillow-layer
mkdir pillow-layer
cd pillow-layer
mkdir python

# Install Pillow compatible with Lambda (Python 3.10)
pip3 install pillow --platform manylinux2014_x86_64 \
--target=python/ \
--implementation cp \
--python-version 3.10 \
--only-binary=:all: \
--upgrade

# Zip the layer
zip -r pillow-layer.zip python

---

Step 3: Download ZIP

• Click Actions → Download file
• Enter:

pillow-layer.zip

---

Step 4: Upload Layer to AWS Lambda

1. Go to Lambda → Layers → Create layer
2. Name: pillow-layer
3. Upload: pillow-layer.zip
4. Runtime: Python 3.10
5. Create

---

Step 5: Attach Layer to Function

• Open Lambda function
• Scroll → Layers
• Remove old layer (if any)
• Add pillow-layer

---

🧪 Example

Input:

Stack_Output.png (79.3 KB)

Output:

resized-Stack_Output.png (10.8 KB)

👉 Significant size reduction improves performance.

---

🔗 Sample Output (CloudFront URL)

https://<your-cloudfront-domain>/resized-image.png

---

🔐 Security

• S3 buckets are not public
• Access is restricted via CloudFront Origin Access Control (OAC)
• IAM roles follow least privilege principle

---

💡 Use Cases

• Social media apps (profile pictures)
• E-commerce platforms (product thumbnails)
• Portfolio websites (image optimization)
• Blogging platforms (faster load times)

---

📈 Advantages

• No server management
• Automatically scales with usage
• Low cost (pay-per-use)
• Faster content delivery worldwide

---

🔥 Future Enhancements

• Multiple image sizes (thumbnail, medium, large)
• Convert images to WebP format
• Add watermarking
• API-based upload using API Gateway
• Frontend UI for uploading images
• Logging and monitoring with CloudWatch

---

🎓 Learning Outcomes

• Understanding of serverless architecture
• Hands-on experience with AWS services
• Event-driven system design
• CDN integration for performance optimization

---

🏁 Conclusion

This project demonstrates how modern cloud applications can be built using serverless technologies. It showcases scalability, automation, and performance optimization using AWS services.

---

👨‍💻 Author

Jai Dev B.Tech CSE | Cloud Computing Enthusiast | Cloud Engineering | DevOps

---