Happy 20th Birthday Amazon S3 The Storage Service That Built the Modern Cloud.

From 5GB Files to 500 Trillion Objects: Celebrating 20 Years of Amazon S3

On March 14, 2026, AWS published a retrospective blog post celebrating the two-decade journey of Amazon S3 (Simple Storage Service), which officially launched on this day in 2006. What started as a simple utility for developers has evolved into the foundational bedrock of the modern internet.

The Early Days of Cloud Storage

When S3 first arrived in 2006, it was a modest service with a 5GB file size limit and a price tag of $0.20 per GB per month. In its infancy, it supported protocols like SOAP, REST, and even BitTorrent, offering developers a new way to store and retrieve data at scale.

Two Decades of Radical Transformation

Twenty years later, the landscape has shifted dramatically. The cost of storage on S3 has plummeted by 85%, now sitting at just $0.023 per GB per month. S3 has also become the core engine powering advanced services, including:

• 2016: Launch of Amazon Athena for serverless queries.

• Recent Years: Introduction of S3 Tables (based on Apache Iceberg) and S3 Vector to support high-performance AI and machine learning workloads.

Today, S3 hosts an incredible 500 trillion objects and handles over 200 million API requests per second.

The Pre-Cloud Origins

While S3 is often cited as the first true "cloud" service from AWS, its roots go back even further. The journey began in 2000 with Merchant.com, a platform allowing retailers to build their own web stores. This was followed by the 2002 launch of Amazon.com Web Services, which provided developers with APIs to access data directly from Amazon.com.

One of the things that made S3 an industry standard is its 99.999999999% (11 Nines) resilience. This means that if you store 10,000 files, you're likely to lose only one file every 10 million years. This level of reliability is what makes global organizations and governments trust S3 to store their critical data.

Before 2006, startups had to invest heavily in purchasing their own data storage hardware. S3 transformed its business model into a "pay-as-you-go" model, reducing financial barriers and allowing companies like Netflix, Pinterest, and Airbnb to scale rapidly without worrying about infrastructure.

In this AI era, S3 is no longer just a place to store junk files; it has become a massive data lake. Features like S3 Express One Zone make retrieving data for AI model training up to 10 times faster than before, a key factor in developing current generative AI technologies.

Today, S3's API has become the universal language for data storage. Even competitors or on-premise storage hardware manufacturers need to make their systems "S3 Compatible" in order to work with software across the world.

 

Adobe Settles FTC Lawsuit for $150M Over "Difficult-to-Cancel" Subscriptions. 

 

Source: AWS 

💬 AI Content Assistant

Ask me anything about this article. No data is stored for your question.

From 5GB Files to 500 Trillion Objects: Celebrating 20 Years of Amazon S3

On March 14, 2026, AWS published a retrospective blog post celebrating the two-decade journey of Amazon S3 (Simple Storage Service), which officially launched on this day in 2006. What started as a simple utility for developers has evolved into the foundational bedrock of the modern internet.

The Early Days of Cloud Storage

When S3 first arrived in 2006, it was a modest service with a 5GB file size limit and a price tag of $0.20 per GB per month. In its infancy, it supported protocols like SOAP, REST, and even BitTorrent, offering developers a new way to store and retrieve data at scale.

Two Decades of Radical Transformation

Twenty years later, the landscape has shifted dramatically. The cost of storage on S3 has plummeted by 85%, now sitting at just $0.023 per GB per month. S3 has also become the core engine powering advanced services, including:

• 2016: Launch of Amazon Athena for serverless queries.

• Recent Years: Introduction of S3 Tables (based on Apache Iceberg) and S3 Vector to support high-performance AI and machine learning workloads.

Today, S3 hosts an incredible 500 trillion objects and handles over 200 million API requests per second.

The Pre-Cloud Origins

While S3 is often cited as the first true "cloud" service from AWS, its roots go back even further. The journey began in 2000 with Merchant.com, a platform allowing retailers to build their own web stores. This was followed by the 2002 launch of Amazon.com Web Services, which provided developers with APIs to access data directly from Amazon.com.

One of the things that made S3 an industry standard is its 99.999999999% (11 Nines) resilience. This means that if you store 10,000 files, you're likely to lose only one file every 10 million years. This level of reliability is what makes global organizations and governments trust S3 to store their critical data.

Before 2006, startups had to invest heavily in purchasing their own data storage hardware. S3 transformed its business model into a "pay-as-you-go" model, reducing financial barriers and allowing companies like Netflix, Pinterest, and Airbnb to scale rapidly without worrying about infrastructure.

In this AI era, S3 is no longer just a place to store junk files; it has become a massive data lake. Features like S3 Express One Zone make retrieving data for AI model training up to 10 times faster than before, a key factor in developing current generative AI technologies.

Today, S3's API has become the universal language for data storage. Even competitors or on-premise storage hardware manufacturers need to make their systems "S3 Compatible" in order to work with software across the world.

 

Adobe Settles FTC Lawsuit for $150M Over "Difficult-to-Cancel" Subscriptions. 

 

Source: AWS