Amazon Revolutionizes Cloud Storage: S3 Buckets Now Function as High-Performance File Systems

Breaking News: AWS Unveils S3 Files – Object Storage Meets File System Flexibility

AWS today announced Amazon S3 Files, a groundbreaking feature that transforms Amazon Simple Storage Service (S3) buckets into fully functional, high-performance file systems. This eliminates the long-standing trade-off between object storage's cost and durability and a file system's interactive, low-latency capabilities.

“For over a decade, customers had to choose between two storage paradigms. Now, they don’t have to,” said Jeff Barr, AWS Chief Evangelist, in an exclusive statement. “S3 Files lets you access any general-purpose bucket as a native file system on EC2, ECS, EKS, or Lambda functions – with automatic synchronization and fine-grained control.”

The new service supports all NFS v4.1+ operations, enabling create, read, update, and delete actions directly on S3 objects as files and directories. Changes made via the file system are instantly reflected in the S3 bucket, making data sharing across compute clusters seamless and eliminating data duplication.

Background: The Storage Divide

Since the dawn of cloud computing, object storage (like Amazon S3) and file systems served different purposes. Object storage excels at massive scale, low cost, and high durability, but lacks the interactivity of a file system. File systems offer low-latency, byte-level modifications but are harder to scale and more expensive for long-term retention.

AWS trainers and architects spent years explaining this distinction, often using analogies – “S3 objects are like books in a library: you can’t edit a page; you replace the whole book.” S3 Files now bridges that gap, offering the best of both worlds without compromise.

How It Works

Under the hood, S3 Files leverages Amazon Elastic File System (EFS) technology combined with intelligent caching and pre-fetching. Frequently accessed data is stored on high-performance local storage for low-latency access. For sequential reads or large files, data is served directly from S3 to maximize throughput.

The system also supports byte-range reads, transferring only requested bytes to minimize data movement and costs. Administrators can fine-tune caching policies – choosing to load full file data or just metadata – to optimize for specific access patterns.

What This Means for Enterprises and Developers

S3 Files unlocks new possibilities for AI, machine learning, and big data workflows. Training ML models often requires fast, interactive access to datasets that traditionally resided in file systems. Now, those datasets can live in S3 while being accessed as files, reducing data silos and operational overhead.

“This is a game-changer for data-intensive applications,” added Barr. “Whether you're running production microservices, training generative AI models, or building agentic systems, you no longer need to copy data between storage tiers. S3 becomes the single source of truth.”

Key Benefits at a Glance

• Seamless integration with all AWS compute resources (EC2, ECS, EKS, Lambda).
• No data duplication – file system changes sync automatically to S3.
• NFS v4.1 support for full CRUD operations.
• Intelligent caching and pre-fetching for low-latency access.
• Fine-grained control over caching policies and metadata.

S3 Files is available today in all commercial AWS Regions. Pricing is based on storage class, data access, and compute resource usage – consistent with existing S3 and EFS pricing models.

Industry analysts predict this will accelerate adoption of S3 as the central data lake for enterprises, while reducing the need for separate file storage systems. “It’s the end of the trade-off,” said Barr. “File system performance with object storage economics.”