AWS Ships S3 Files

For nearly 20 years, Amazon Simple Storage Service was the place you dumped huge piles of data, but not the place most old file-based software could work directly. On April 7, 2026, Amazon Web Services changed that by letting a bucket show up like a normal shared file system. (aws.amazon.com) Cloud storage has usually come in two camps. Object storage is like a warehouse of sealed boxes you fetch by name, while a file system is like an office filing cabinet where programs expect folders, locks, renames, and shared editing. (docs.aws.amazon.com) That mismatch created a lot of ugly plumbing. Companies kept one copy of data in Amazon Simple Storage Service for cheap scale and another copy in a separate file system so older analytics tools, machine learning jobs, and internal apps could open files the way they were written to. (aws.amazon.com) Amazon S3 Files is Amazon Web Services’ answer to that split. It gives Amazon Simple Storage Service full file-system behavior on top of existing buckets, so applications can read, write, rename, and organize data as files while the original data stays in Amazon Simple Storage Service. (docs.aws.amazon.com) Amazon says the service is built using Amazon Elastic File System, which is its managed network file system. In practice, that means Amazon S3 Files borrows a fast caching layer for small, latency-sensitive work, then syncs changes back to the bucket underneath. (docs.aws.amazon.com) The speed trick is selective caching, not moving your whole bucket into expensive storage. By default, files up to 128 kilobytes get pulled into the high-performance layer when latency matters most, while many larger reads go straight to Amazon Simple Storage Service, including reads of 1 megabyte or more after data has synced. (docs.aws.amazon.com) Amazon Web Services also says thousands of compute resources can attach to the same Amazon S3 file system at once. That matters for clusters running on Amazon Elastic Compute Cloud, Amazon Elastic Kubernetes Service, Amazon Elastic Container Service, or Amazon Web Services Lambda, which are the four compute services the documentation lists for mounting it. (aws.amazon.com) (docs.aws.amazon.com) The pitch is “no migration required,” which is the most enterprise part of this launch. Amazon says new and existing bucket data works as-is, so teams do not have to rewrite applications around object-storage application programming interfaces just to get into the cloud. (aws.amazon.com) This also lands in the middle of Amazon’s push to make Amazon Simple Storage Service do more than hold backups and archives. In the last two years, Amazon Web Services has added products like Amazon S3 Tables for analytic table data and now Amazon S3 Files for file-style access, turning the bucket into a base layer for more kinds of workloads. (aws.amazon.com) (techtalkreplay.com) The competitive angle is straightforward. File-storage vendors like NetApp, Qumulo, and VAST Data have been selling ways to bridge file and object storage for years, and now the biggest cloud object store is building more of that bridge itself. (blocksandfiles.com) Amazon S3 Files is generally available in 34 Amazon Web Services Regions as of launch. If it works the way Amazon promises, a lot of “copy the data into a file share first” jobs may quietly disappear from cloud architecture diagrams this year. (aws.amazon.com)