Virtual drives that make advanced storage feel native.

FUSION VFS · VIRTUAL FILE SYSTEMFusion VFS puts virtual drives in the Windows file browser. Files and applications work as they always do, while every file is stored with Fusion's content-defined deduplication, compression and encryption underneath. It is also the compression and deduplication built into Fusion OS.

See how it works How files are stored

Exact deduplication

Repeated data is stored once, identified by SHA-256.

Content-aware compression

Brotli where it helps, raw where it does not.

Encrypted at rest

Every chunk is sealed with AES-256-GCM.

Native files

A drive letter in the file browser, not a storage portal.

A drive letter, not a new app to learn.

Mount a Fusion VFS repository as a drive and keep using the file browser and the applications you already have. Browse folders, open, save, rename and delete files through normal operating-system calls while Fusion VFS manages the storage below them.

Fusion VFS for Windows

Virtual drives appear and behave like any other drive in the Windows file browser, mounted through WinFsp.

Transparent to applications

Range reads, sparse writes, flushes, directory listings, attributes, timestamps and rename all map to Fusion VFS, so applications need no changes.

Windows file behavior

Directories, Windows attributes, timestamps and stable file IDs persist, renames keep a file's identity, and open handles follow Windows read, write and delete sharing rules.

Every file, stored efficiently and securely.

Fusion VFS represents each file as a small manifest. Its data is split at content-defined boundaries, so an insert or edit changes only the chunks around it, and each chunk is deduplicated, compressed when that saves space, and encrypted.

Content-defined chunking

Chunk boundaries follow the data rather than fixed offsets, so similar files and new versions share most of their chunks.

Deduplication by scope

Chunks are deduplicated by SHA-256 within a scope. Use separate scopes wherever data must not deduplicate across security domains.

Compression that pays its way

Brotli compression is applied only where it makes a chunk smaller, so already-compressed media is not slowed down for nothing.

AES-256-GCM encryption

Every stored chunk is encrypted with the repository key, which you keep and back up separately.

Seekable reads

Opening or seeking a file decrypts and decompresses only the chunks a read touches, never the whole file.

More space at high speed

Removing repeated data raises effective capacity while parallel work and fast access stay intact.

Writes that survive a crash.

Open files are written to a sparse staging view, and a checkpoint commits them in one atomic step, so readers always see a complete version.

Sparse staging

Random writes and size changes touch only the dirty ranges of a file; unchanged data is read from the last committed version.

Atomic checkpoints

On flush or close, one content-defined pass writes only the missing chunks and swaps the file's manifest atomically.

Recovery after a stop

If the process or machine stops before close, flushed changes are recovered onto the last committed version the next time the drive is mounted.

One storage foundation across Fusion.

Fusion VFS uses the same storage format as Fusion's backup and storage products, so content identity, deduplication, compression and integrity work the same everywhere.

Fusion OS

Fusion VFS is the compression and deduplication built into Fusion OS, shared by everything that stores files there.Fusion OS →

Vault

Vault Backup Client and Server, Vault Cloud Storage and the VaultX storage appliances build on the same foundation.fusionvault.cloud →

Solerra

Fusion's language and runtime connect storage to applications through tested contracts.solerra.cloud →

Fusion VFS is in active development.

Follow Fusion VFS and the rest of the Fusion portfolio on fusionsoftware.cloud.Fusion VFS on fusionsoftware.cloud