July 1st 2026
FiberFS is officially available. FiberFS is a POSIX filesystem built on S3 compatible storage. Why use S3 for a filesystem? First, S3 storage is ubiquitous. Implementations are everywhere, both in the cloud and on premise. S3 is steadily becoming a commodity storage product and the de facto storage API, which makes it hard to ignore. Second, S3 has proven to be performant, reliable, and scalable. Its simple key/value based design can be easily distributed across a multitude of servers and physical disks. So unlike most filesystems which have the complexity of disk management, FiberFS has the advantage of neatly abstracting physical disks away into a system that is well understood and easily supported.
So FiberFS is just an S3 compatible filesystem? Not quite, it offers a lot more than just a filesystem view of S3. First, it's a POSIX compatible networked filesystem. It has a custom distributed filesystem protocol built for S3 and it supports a virtually unlimited number of concurrent readers and writers across any number of hosts. Second, unlike other distributed filesystems, it doesn’t use a database or any 3rd party products, it can do everything it needs using just a plain S3 API. Finally, it’s 100% CDN compatible. Anything can be cached anywhere for any amount of time. FiberFS has been designed with the capability to do a lot more than what has typically been expected of S3 based filesystems and it’s ready to compete with some of the most advanced distributed filesystems available in terms of features, simplicity, and performance.