Skip to main content


FlexFS is an innovative file system created by Paradigm4 that can easily transform any network-accessible key-value store with get, put, delete, list operations into a POSIX-compliant network file system which can be mounted across several hosts concurrently. A flexFS volume, when mounted on a host, appears to an end-user as an ordinary folder. FlexFS can be used in a variety of environments: on-premise, in the cloud, or in a hybrid configuration that straddles both. In practice, flexFS is typically backed by an object storage system (e.g. S3, Google Cloud Storage, Azure Blob, Wasabi, MinIO, Scality) and derives performance characteristics of that sytem (e.g. throughput, durability, latency, cost).

FlexFS supports all POSIX-compliant tools and utilities and makes it easy to build reliable, portable, and performant workflows that operate against files in flexFS as if they were located on the host executing the workflow, even though file data will automatically stream to and from the host on-demand. FlexFS automatically handles the difficult and error-prone work of dealing with low-level details such as transient failures, compression, caching, and encryption.

In general, flexFS is a drop-in replacement for general-purpose, POSIX-compliant file systems offered by cloud providers (e.g Amazon EFS, FSx for Lustre) and other vendors. One notable difference, however, is that flexFS is designed to provide high elastic throughput by default, and at a low cost-point, whereas most competing file systems provide low-latency by default and charge extra to add throughput capacity. This difference has major cost and time-to-answer implications when performing large-scale analyses, working with large files, and sharing file volumes with many users at once. And those are the realms where flexFS demonstrates dominance.



Additional background information can be found in the following blog articles: