Chapter 42: AWS S3

AWS S3

If AWS is a huge supermarket of cloud tools, then S3 is the biggest, most popular aisle — the one where almost every company (Zomato, Swiggy, PhonePe, Byju’s, Unacademy, JioCinema, Hotstar, and thousands of Hyderabad startups) keeps their photos, videos, backups, logs, static websites, data lakes, ML datasets, invoices, user uploads… basically almost every file or object that isn’t a live database or attached disk.

Let me explain it like we’re sitting together with a whiteboard and a second cup of Irani chai — slow, detailed, full of real-life analogies, actual Hyderabad examples, 2026 reality, pricing, storage classes, security best practices, and a simple hands-on you can do right now.

1. What is Amazon S3? (Very Simple First)

Amazon S3 is a fully managed object storage service that lets you store and retrieve any amount of data (from a few bytes to exabytes) at any time, from anywhere on the internet.

• You store objects — each object = a file + metadata + unique key (like a filename in a folder)
• There is no practical limit to how much data you can store or how many objects
• Data is automatically replicated across multiple devices and (optionally) multiple Availability Zones → 99.999999999% (11 9s) durability
• You pay only for what you actually use (storage + requests + data transfer)

Official short line (still perfect in 2026): “Amazon S3 provides object (file) storage through a web service interface. It is built to store and retrieve any amount of data from anywhere.”

In plain Hyderabad language: Imagine S3 as an infinite, indestructible, super-secure godown (warehouse) somewhere in the cloud. You can throw anything inside (photos, videos, PDFs, JSON files, backups, Excel sheets, entire website folders…) You give each item a unique name (key) like “user123/profile-photo-2026.jpg” You can lock it, encrypt it, version it, make it public, serve it via CDN, run code when someone touches it… and it never gets full, never breaks, and never loses anything.

2. Why Almost Every Hyderabad Company Uses S3 (Real Reasons 2026)

• Unlimited scale — store 1 GB or 1 petabyte — same API, same price model
• Durability — 11 9s = statistically you lose one object in ~10 million years
• Availability — 99.99% SLA for Standard class (higher with multi-AZ options)
• Cheap — especially in India Regions (ap-south-1 & ap-south-2)
• Versatile — static website hosting, big data lake, ML datasets, backups, logs, media streaming…
• Event-driven — S3 Event Notifications → trigger Lambda, SQS, SNS when file uploaded/deleted
• Security & compliance — encryption, versioning, object lock (immutable), access logging, bucket policies, IAM, VPC endpoints, MFA delete…

Real Hyderabad examples (2026):

• Zomato/Swiggy → millions of restaurant photos, menu PDFs, user-uploaded food pics
• PhonePe/Google Pay → transaction receipts, KYC documents (encrypted + immutable)
• Byju’s/Unacademy → course videos, PDFs, recorded lectures (served via CloudFront)
• Your college friend’s startup → user profile photos, short videos, daily backups
• JioCinema/Hotstar → streaming video chunks (origin in S3 + CloudFront)
• ISRO / government projects → satellite imagery archives

3. Core Concepts of S3 (The Building Blocks)

4. S3 Storage Classes – Quick 2026 Guide (Choose Wisely!)

2026 tip for Hyderabad teams: Start with Intelligent-Tiering on most buckets — it automatically saves money when data gets cold, no manual lifecycle rules needed.

5. Real Hands-On Example – Your First S3 Bucket (Do This Today!)

1. Go to S3 console → Create bucket
   • Name: “my-hyderabad-first-bucket-2026” (globally unique!)
   • Region: ap-south-2 (Asia Pacific – Hyderabad)
   • Keep Block all public access (safe default)
2. Upload a photo or selfie → name it “profile.jpg”
3. Make it public (for learning only!):
   • Bucket → Permissions → Block public access → edit → uncheck
   • Bucket policy → paste:
     JSON

     PHP

     0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

     {   "Version": "2012-10-17",   "Statement": [     {       "Effect": "Allow",       "Principal": "*",       "Action": "s3:GetObject",       "Resource": "arn:aws:s3:::my-hyderabad-first-bucket-2026/*"     }   ] }

4. Copy object URL → open in browser → your photo is now hosted on AWS!

Cost? Usually ₹0–5 for this experiment (first 5 GB free, first 20,000 GET requests free).

Summary Table – S3 Cheat Sheet (2026 – India Focus)

Teacher’s final note: S3 is the “infinite, indestructible hard disk in the sky” that almost every serious application in Hyderabad touches every day — photos, videos, backups, logs, static sites, data lakes… Learn S3 first, master buckets, keys, policies, lifecycle, and CloudFront → you can build 70% of real-world apps.

Got it? This is the “where do I actually put all my files?” lesson.

Next?

• Deep dive on S3 storage classes + lifecycle + cost optimization?
• How to host a static React website on S3 + CloudFront?
• Or S3 event → Lambda thumbnail generator full example?

Tell me — next whiteboard ready! 🚀📦