Blogs by Jay Tillu

How Amazon S3 Powers Modern Cloud Storage

·

5 min read

Cover Image for How Amazon S3 Powers Modern Cloud Storage

Meet Arjun — an ambitious developer building a startup on AWS. One day, he needed a place to store:

  • User uploads

  • Log files

  • Backups

  • Website images

  • Even entire static websites

That’s when his mentor said:

“You need Amazon S3 — the infinitely scalable cloud file cabinet.”

So Arjun started his journey to learn everything about Amazon Simple Storage Service (S3) — a service that powers websites, backups, mobile apps, data lakes, and more.


🧱 What Is Amazon S3?

Amazon S3 (Simple Storage Service) is AWS’s object storage service that lets you store and retrieve files (objects) at virtually unlimited scale.

🧠 Key Concepts:

  • ✅ Infinitely scalable (you can store unlimited data)

  • ✅ Durable and highly available (99.999999999% durability — that’s 11 9’s!)

  • ✅ Pay-as-you-go pricing

  • ✅ Secure, fast, and integrated with other AWS services

It’s like Dropbox, but for developers and enterprise-level cloud apps.


📦 Buckets: S3’s Top-Level Containers

Everything in S3 is stored inside a bucket. Think of a bucket like a top-level folder in your storage.

Important Rules About Buckets:

  • Must have a globally unique name

  • Created in a specific AWS region

  • Maximum: 100 buckets per AWS account (by default)

  • S3 looks like a global service but buckets are created in a region

  • Names:

    • Lowercase letters only (no uppercase or underscores)

    • 3 to 63 characters

    • No IP address-style names (like 192.168.0.1)

    • Must NOT start with the prefix xn --

    • Must NOT end with the suffix -s3alias

📘 SAA Tip: Bucket names must be unique across all AWS accounts and regions.


📄 Objects: The Files Inside Buckets

An object is the actual file or data you store in S3. It can be:

  • Text file

  • Image

  • Video

  • Backup archive

  • Anything else (up to 5TB in size!)

Object Anatomy:

ComponentDescription
KeyThe full path of the object (like folder1/folder2/myfile.txt)
ValueThe file's contents (binary or text data)
MetadataExtra info like file type, custom labels
TagsKey-value labels for organization & automation
Version IDIf versioning is enabled, each change gets a unique version

📘 SAA Tip: S3 doesn’t use real folders — “folders” are simulated using the object key prefix (e.g., images/cat.jpg).


📂 Object Keys vs. Folders

S3 doesn’t have traditional folders. Instead:

  • Keys are the full path of the file (e.g., backup/logs/2024/report.csv)

  • The "folders" you see in the console are just part of the key

✅ There’s no actual folder structure — it’s just a long string with slashes (/).


📤 Uploading Big Files: Multipart Upload

  • You can upload files up to 5TB

  • For any file larger than 5GB, you must use Multipart Upload

    • Breaks your file into smaller chunks (parts)

    • Uploads each part in parallel

    • Faster, more reliable for large uploads

📘 Example: Uploading a 2TB video → split into 400 parts of 5GB each


🧾 Metadata, Tags, and Versioning

🏷️ Metadata

  • Info like content type (image/png, application/pdf)

  • Can be user-defined or system-generated

🔖 Tags

  • Up to 10 key-value pairs per object

  • Useful for:

    • Billing (tag by team or project)

    • Lifecycle policies

    • Permissions

🕓 Versioning

  • If enabled, every update or deletion creates a new version

  • Helps protect against accidental deletion or overwrite

📘 SAA Tip: Even deleted objects are soft-deleted when versioning is turned on.


📂 Common S3 Use Cases

Use CaseDescription
Backup & RestoreStore snapshots, databases, VM images
Disaster RecoveryReplicate to another region for resilience
Static Website HostingServe HTML, CSS, JS with no backend
Big Data & AnalyticsStore structured/unstructured data for Athena, EMR
Media HostingVideos, images, podcasts for streaming or download
Software DistributionDeliver app updates, patches, installers
Hybrid StorageExtend on-premises storage into AWS

🧊 S3 Storage Classes (Intro Only)

Different storage classes = different pricing + durability tradeoffs:

Storage ClassUse Case
StandardFrequent access (default)
Infrequent Access (IA)Rarely accessed, but fast retrieval
GlacierArchive, very cheap, slow to restore
Glacier Deep ArchiveCheapest, long-term cold storage
Intelligent-TieringAuto-moves data between hot/cold

📘 Real-World Examples

  • 🧾 NASDAQ stores 7 years of trade data in S3 Glacier

  • 📊 Sysco runs data analytics on their S3 data lake

Even Netflix, Reddit, and NASA rely on Amazon S3!


📌 Summary — Key Points for AWS SAA

FeatureDescription
S3Object storage with near-infinite scale
BucketTop-level container, globally unique
ObjectFile + key + metadata + tags
Max Object Size5TB (use Multipart Upload if > 5GB)
No real foldersKeys simulate folders with slashes
VersioningProtects against accidental deletes
Use CasesBackup, hosting, analytics, disaster recovery

🧠 Arjun’s Final Takeaway

“Amazon S3 isn’t just a file cabinet — it’s the spine of modern cloud apps. Whether you’re storing PDFs or petabytes, it scales, protects, and powers your data.”


More AWS SAA Articles

Follow me for more such content