rt/mergerfs
1
0
Fork 0
mirror of https://github.com/trapexit/mergerfs.git synced 2026-10-25 18:40:18 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

Improve proj comparisons + new SnapRAID details

Browse source
This commit is contained in:
Antonio SJ Musumeci 2026-03-31 13:00:19 -05:00 • committed by trapexit
parent b8776d205d
commit ecee84b323
2 changed files with 363 additions and 200 deletions
Download patch file Download diff file Expand all files Collapse all files

41
mkdocs/docs/faq/compatibility_and_integration.md
View file

@ -67,7 +67,38 @@ Yes. They are completely unrelated pieces of software that just happen
to work well together.
## Can I use mergerfs without nonraid? nonraid without mergerfs?
## How does mergerfs interact with SnapRAID recovery behavior?
SnapRAID does not choose parity layout based on directory structure or
relative path. As each file is discovered, SnapRAID fills parity space
for that file based on its size using the next available parity area.
This makes the logical parity layout depend on discovery order, file
size, and how much data already exists on each filesystem at that
time. Filesystems with similar used space, regardless of total
capacity, are more likely to place newly added files into similar
parity positions.
Because of that, if files are changed or removed across multiple
branches and those files happen to overlap in parity space, recovery
of those deleted files may be less reliable. This is not a general
concern, is not unique to mergerfs usage, and it does not affect
filesystem failure recovery. It is specifically a concern when trying
to recover from accidental deletion of files spread across multiple
filesystems.
In practice this mostly matters when the filesystems have very similar
used space, the files are created across those filesystems around the
same time, and you later want SnapRAID to recover an accidental
deletion spanning multiple filesystems. Otherwise this is usually not
something to worry about.
If, however, this is a concern it can be beneficial to keep files on
as few filesystems as possible or otherwise keep related files
together. See the [collocation
section.](configuration_and_policies.md#how-can-i-ensure-files-are-collocated-on-the-same-branch)
## Can I use mergerfs without NonRAID? NonRAID without mergerfs?
[https://github.com/qvr/nonraid](https://github.com/qvr/nonraid)
@ -75,6 +106,14 @@ Yes. They are completely unrelated pieces of software that just happen
to work well together.
### How does mergerfs interact with NonRAID?
There are no known special interactions or concerns between mergerfs
and NonRAID. They are separate tools and mergerfs has no known
behavior which requires any special configuration or caveats when used
with NonRAID.
## Does mergerfs support CoW / copy-on-write / writes to read-only filesystems?
Not in the sense of a filesystem like BTRFS or ZFS nor in the

522
mkdocs/docs/project_comparisons.md
View file

@ -7,14 +7,19 @@ There are no solutions, only trade-offs.
* [https://romanrm.net/mhddfs](https://romanrm.net/mhddfs)
mhddfs is, like mergerfs, a FUSE based filesystem which can pool
mhddfs is, like mergerfs, a FUSE-based filesystem which can pool
multiple filesystems together.
mhddfs had not been updated in over a decade and has known stability
and security issues. mergerfs provides a super set of mhddfs' features
and offers better performance. As of 2020 the author of mhddfs has
and security issues. mergerfs provides a superset of mhddfs' features
and offers better performance. As of 2020, the author of mhddfs has
[moved to using mergerfs.](https://romanrm.net/mhddfs#update)
* **Better than mergerfs for:** legacy setups that already depend on
mhddfs-specific behavior and do not need newer features.
* **Worse than mergerfs for:** most modern deployments due to maintenance
status, feature coverage, and reliability.
Below is an example of mhddfs and mergerfs setup to work similarly.
```
@ -29,48 +34,73 @@ mergerfs -o minfreespace=4G,category.create=ff /mnt/drive1:/mnt/drive2 /mnt/pool
* [https://aufs.sourceforge.net](https://aufs.sourceforge.net)
* [https://en.wikipedia.org/wiki/Aufs](https://en.wikipedia.org/wiki/Aufs)
aufs, another union filesystem, is a kernel based overlay filesystem
aufs, another union filesystem, is a kernel-based overlay filesystem
with basic file creation placement policies.
Compared with mergerfs, aufs focuses on kernel-level overlay and
copy-on-write behavior, while mergerfs focuses on userspace pooling of
existing filesystems with richer branch placement policies.
aufs failed to be included in the mainline kernel and is no longer
available in most Linux distros, making it harder to get installed for
the average user. Development has been largely dormant for years.
While aufs can often offer better peak performance due to being
primarily kernel based (at least when `passthrough.io` is disabled in
primarily kernel-based (at least when `passthrough.io` is disabled in
mergerfs), mergerfs provides more configurability and is generally
easier to use. mergerfs however does not offer the overlay /
easier to use. mergerfs, however, does not offer the overlay /
copy-on-write (CoW) features which aufs has.
* **Better than mergerfs for:** kernel-level overlay/CoW workloads where
that behavior is required.
* **Worse than mergerfs for:** users needing a widely packaged,
actively-maintained, easy-to-deploy pooling solution.
## unionfs
* [https://unionfs.filesystems.org](https://unionfs.filesystems.org)
unionfs for Linux is a "stackable unification file system" which
functions like many other union filesystems. unionfs is no longer
maintained and was last released for Linux v3.14 back in 2014.
functions like many other union filesystems. It was last released for
Linux v3.14 back in 2014.
Compared with mergerfs, unionfs is an older kernel union filesystem
implementation while mergerfs is a modern, actively-maintained
FUSE-based pooling filesystem with broader policy controls.
Documentation is sparse so a comparison of features is not possible
but given the lack of maintenance and support for modern kernels there
is little reason to consider it as a solution.
* **Better than mergerfs for:** historical or legacy environments that
already depend on this specific kernel unionfs behavior.
* **Worse than mergerfs for:** practical modern deployments due to
limited maintenance, aging kernel support, and sparse documentation.
## unionfs-fuse
* [https://github.com/rpodgorny/unionfs-fuse](https://github.com/rpodgorny/unionfs-fuse)
unionfs-fuse is more like unionfs, aufs, and overlayfs than mergerfs
in that it offers overlay / copy-on-write (CoW) features. If you're
just looking to create a union of filesystems and want flexibility in
file/directory placement then mergerfs offers that whereas
unionfs-fuse is more for overlaying read/write filesystems over
read-only ones.
unionfs-fuse is a FUSE-based union filesystem focused on overlay
behavior, including copy-on-write (CoW) style layering of writable
content over read-only content.
Since unionfs-fuse, as the name suggests, is a FUSE based technology
Compared with mergerfs, unionfs-fuse's primary feature is
overlay/CoW-style layering, while mergerfs' primary feature is
policy-based pooling and placement across multiple regular writable
filesystems.
Since unionfs-fuse, as the name suggests, is a FUSE-based technology
it can be used without elevated privileges that kernel solutions such
as unionfs, aufs, and overlayfs require.
* **Better than mergerfs for:** overlay-style unions with copy-on-write
semantics in userspace.
* **Worse than mergerfs for:** flexible create-policy-based pooling of
many writable filesystems.
## overlayfs
@ -78,12 +108,18 @@ as unionfs, aufs, and overlayfs require.
overlayfs is effectively the functional successor to unionfs,
unionfs-fuse, and aufs and is widely used by Linux container platforms
such as Docker and Podman. Both overlayfs and FUSE were originally created by Miklos Szeredi.
such as Docker and Podman. Both overlayfs and FUSE were originally
created by Miklos Szeredi.
If your use case is layering a writable filesystem on top of read-only
filesystems then you should look first to overlayfs. Its feature set
however is very different from mergerfs and solves different problems.
* **Better than mergerfs for:** container and image-layering workflows
where copy-on-write semantics are central.
* **Worse than mergerfs for:** heterogeneous pooling use cases that need
flexible branch selection policies across many existing filesystems.
## RAID0, JBOD, SPAN, drive concatenation, striping
@ -92,23 +128,27 @@ however is very different from mergerfs and solves different problems.
* [SPAN](https://en.wikipedia.org/wiki/Non-RAID_drive_architectures#Concatenation_(SPAN,_BIG))
* [striping](https://en.wikipedia.org/wiki/Data_striping)
These are block device technologies which in some form aggregate
devices into what appears to be a singular device on which a
traditional filesystem can be used. The filesystem has no
understanding of the underlying block layout and should one of those
underlying devices fail or be removed the filesystem will be missing
that chunk which could contain critical information and the whole
filesystem may become unrecoverable. Even if the data from the
filesystem is recoverable it will take using specialized tooling to do
so.
These are block device technologies which, in some form, aggregate
devices into what appears to be a single device on which a traditional
filesystem can be used. The filesystem has no understanding of the
underlying block layout and, should one of those underlying devices
fail or be removed, the filesystem will be missing that chunk, which
could contain critical information, and the whole filesystem may
become unrecoverable. Even if the data from the filesystem is
recoverable, doing so may require specialized tooling.
In contrast, with mergerfs you can format devices as one normally
would or take existing filesystems and then combine them in a pool to
In contrast, with mergerfs you can format devices as you normally
would, or take existing filesystems and combine them in a pool to
aggregate their storage. The failure of any one device will have no
impact on the other devices. The downside to mergerfs' technique is
the fact you don't actually have contiguous space as large as if you
used those other technologies. Meaning you can't create a file greater
than 1TB on a pool of two 1TB filesystems.
impact on the other devices. The downside to mergerfs' approach is
that you do not actually have contiguous space as large as you would
with those other technologies. That means you cannot create a file
larger than 1TB on a pool of two 1TB filesystems.
* **Better than mergerfs for:** creating one contiguous block device and
maximizing single-file size/performance characteristics.
* **Worse than mergerfs for:** failure isolation and straightforward
per-disk recovery when a device dies.
## RAID5, RAID6
@ -117,9 +157,9 @@ than 1TB on a pool of two 1TB filesystems.
* [RAID6](https://en.wikipedia.org/wiki/Standard_RAID_levels#RAID_6)
RAID5, RAID6, and similar technologies aggregate multiple devices at
the block level and offer 1 or more live parity calculations that can
allow the pool to continue to run should 1 device fail and rebuild the
data once the device is replaced.
the block level and provide one or more live parity schemes that can
allow the pool to continue to run should one device fail and rebuild
the data once the device is replaced.
mergerfs offers no parity or redundancy features so in that regard the
technologies are not comparable. [SnapRAID](https://www.snapraid.it)
@ -132,24 +172,100 @@ not done in real-time but NonRAID's are.
For more details and comparison of SnapRAID to related technologies
see [https://www.snapraid.it/compare](https://www.snapraid.it/compare).
* **Better than mergerfs for:** live parity protection and continued array
operation through configured disk failures.
* **Worse than mergerfs for:** incremental mixed-size pooling with simple
per-filesystem independence.
## LVM
* [https://en.wikipedia.org/wiki/Logical_Volume_Manager_(Linux)](https://en.wikipedia.org/wiki/Logical_Volume_Manager_(Linux))
* [https://sourceware.org/lvm2](https://sourceware.org/lvm2)
LVM is a block-level volume management system for Linux. It combines
physical devices into volume groups and creates logical volumes on top,
with features like snapshots, resizing, and thin provisioning.
Compared with mergerfs, LVM operates below the filesystem while mergerfs
operates above filesystems. With LVM, filesystems are created on logical
volumes and do not know about the underlying disks. If a linear or
striped logical volume spans multiple disks and one disk fails, the
filesystem can become unreadable. With mergerfs, each disk typically has
its own independent filesystem, so a single disk failure affects only the
data on that disk.
* **Better than mergerfs for:** block-level features like thin
provisioning, snapshots, and online resizing.
* **Worse than mergerfs for:** incremental mixed-disk pooling where
per-disk failure isolation and per-disk native access are priorities.
## mdadm
* [https://en.wikipedia.org/wiki/Mdadm](https://en.wikipedia.org/wiki/Mdadm)
* [https://raid.wiki.kernel.org/index.php/A_guide_to_mdadm](https://raid.wiki.kernel.org/index.php/A_guide_to_mdadm)
`mdadm` is Linux userspace tooling for configuring and managing software
RAID arrays provided by the kernel `md` driver. It can build RAID0/1/4/5/6/10
and related layouts, monitor arrays, and handle rebuild workflows.
Like RAID technologies discussed above, `mdadm` arrays aggregate devices
at the block layer and present one virtual device to a filesystem. That
can provide performance and/or redundancy depending on RAID level, but it
also means the filesystem depends on the array layout. mergerfs does not
provide block-level RAID; it pools already-formatted filesystems and can
be combined with tools like SnapRAID or NonRAID when redundancy is needed.
* **Better than mergerfs for:** kernel software RAID with real-time
redundancy and array-level performance tuning.
* **Worse than mergerfs for:** flexible mixed-size drive pooling where
per-filesystem independence and simple incremental expansion matter.
## device mapper
* [https://en.wikipedia.org/wiki/Device_mapper](https://en.wikipedia.org/wiki/Device_mapper)
* [https://docs.kernel.org/admin-guide/device-mapper](https://docs.kernel.org/admin-guide/device-mapper)
Device mapper is a kernel framework for creating virtual block devices
from one or more underlying devices. Technologies such as LVM,
`dm-crypt`, `dm-cache`, `dm-thin`, and multipath are built on top of it.
Because device mapper works at the block layer, it is not directly
comparable to mergerfs. It is complementary: you can use device mapper
for block-level functionality (encryption, caching, thin provisioning,
etc.) and then place filesystems on those devices, which mergerfs can
later pool if desired.
* **Better than mergerfs for:** block-layer transformations such as
transparent encryption, caching, and provisioning features.
* **Worse than mergerfs for:** presenting one unified pathname-level view
across many independent, already-formatted filesystems.
## UnRAID
* [https://unraid.net](https://unraid.net)
UnRAID is a full OS and offers a filesystem (likely FUSE based) which
provides a union of filesystems like mergerfs but with the addition of
live parity calculation and storage. Outside parity calculations
mergerfs offers more features and due to the lack of real-time parity
calculation can have higher peak performance. For some users mergerfs
being open source is also preferable.
UnRAID is a full OS platform that offers pooled storage with integrated
real-time parity. Compared with mergerfs, its main advantage is that it
bundles pooling and parity in one product, while mergerfs is a
standalone Linux pooling filesystem without built-in redundancy. For
some users mergerfs' Linux distro-agnostic and fully open-source
approach is also preferable.
For semi-static data mergerfs + [SnapRAID](https://www.snapraid.it)
provides a similar, but not real-time,
solution. [NonRAID](https://github.com/qvr/nonraid) (see below) is a
For semi-static data, mergerfs + [SnapRAID](https://www.snapraid.it)
provides a similar, but not real-time, solution.
[NonRAID](https://github.com/qvr/nonraid) (see below) is a
fork of UnRAID's parity calculation solution and can also be used with
mergerfs.
* **Better than mergerfs for:** turnkey pooling plus integrated real-time
parity in one product.
* **Worse than mergerfs for:** users who prefer a Linux distro-agnostic,
fully open-source component approach.
## NonRAID
@ -162,9 +278,18 @@ single device. Each device gets a virtual block device associated with
it which can be formatted with the filesystem of the user's choice and
will be individually accessible. Similar to SnapRAID but real-time.
Compared with mergerfs, NonRAID adds kernel-level real-time parity while
still allowing per-disk filesystems, whereas mergerfs focuses on pooling
views and placement policies without parity.
Given each device has its own filesystem like a traditional setup it can
be used with mergerfs for those looking for a unified view.
* **Better than mergerfs for:** real-time parity protection without
traditional RAID striping behavior.
* **Worse than mergerfs for:** users who want the simplest setup and do
not need parity, since it adds operational complexity.
## ZFS
@ -175,24 +300,15 @@ snapshots, data integrity checking, and built-in RAID-like redundancy
all in one package. It offers features like transparent compression,
encryption, and automatic data repair.
Unlike mergerfs which focuses on simple pooling of existing
filesystems, ZFS is a complete filesystem that manages storage
directly. ZFS requires more resources and planning to set up, making
it better suited for systems where data protection and integrity are
critical priorities. mergerfs is simpler and lighter weight, making it
ideal for users who want to add storage incrementally without the
overhead of ZFS.
Compared with mergerfs, ZFS is a full storage stack that manages disks
directly with integrated integrity and redundancy features, while
mergerfs is a lightweight FUSE layer that pools existing filesystems
without built-in redundancy.
Both technologies can aggregate multiple drives into a single pool,
but they approach the problem differently. ZFS integrates everything
into the filesystem, while mergerfs pools existing filesystems. For
write-once, read-many workloads like bulk media storage where backup
and data integrity is managed separately, mergerfs is often the better
choice due to lower resource usage and simpler management.
It's [common to see](https://perfectmediaserver.com) both used
together: critical data stored on a ZFS pool with redundancy, while
bulk media is pooled across regular filesystems using mergerfs.
* **Better than mergerfs for:** integrated data integrity, snapshots,
checksumming, and built-in redundancy.
* **Worse than mergerfs for:** low-overhead incremental pooling of
existing filesystems with minimal planning.
## ZFS AnyRAID
@ -201,17 +317,23 @@ bulk media is pooled across regular filesystems using mergerfs.
[ZFS AnyRAID](https://github.com/openzfs/zfs/discussions/16773)
([announcement](https://hexos.com/blog/introducing-zfs-anyraid-sponsored-by-eshtek))
is a feature being developed for ZFS to provide more flexibility in
pools by allowing mixed-capacity disks while maintaining live
redundancy.
is a feature under development for ZFS intended to provide more
flexibility in pools by allowing mixed-capacity disks while
maintaining live redundancy.
If released, ZFS AnyRAID would offer somewhat similar flexibility to
mergerfs' pooling approach, but with built-in redundancy. mergerfs
already provides this flexibility today: add drives of any size at any
time with no redundancy overhead. If you need redundancy with that
flexibility, ZFS AnyRAID could be an option when available; until
then, mergerfs remains the simpler choice for mixed-capacity pooling
with redundancy and integrity available via SnapRAID and/or NonRAID.
If released, ZFS AnyRAID would offer some of the flexibility associated
with mergerfs-style mixed-capacity expansion, but with built-in
redundancy inside ZFS. mergerfs already provides that mixed-size,
add-drives-anytime flexibility today, albeit without built-in
redundancy. If you need redundancy with that flexibility, ZFS AnyRAID
could become an option when available; until then, mergerfs remains the
simpler currently available choice for mixed-capacity pooling, with
redundancy and integrity available via SnapRAID and/or NonRAID.
* **Better than mergerfs for:** users who want mixed-capacity flexibility
and built-in live redundancy in one filesystem stack.
* **Worse than mergerfs for:** users who need mature availability now and
the simplest add-drives-anytime workflow without redundancy overhead.
## Bcachefs
@ -219,78 +341,58 @@ with redundancy and integrity available via SnapRAID and/or NonRAID.
* [https://bcachefs.org](https://bcachefs.org)
Bcachefs is a modern copy-on-write filesystem for Linux designed for
multi-device storage aggregation. It combines the data integrity and
crash-safety features of ZFS with the performance optimization of
high-performance filesystems like XFS and ext4.
multi-device storage aggregation. It includes integrated features such
as checksumming, snapshots, compression, encryption, and redundancy.
Bcachefs is fundamentally a multi-device filesystem. It can aggregate
multiple physical disks of arbitrary sizes into a single logical
filesystem, allowing data to be automatically distributed across
devices. It includes built-in replication capabilities similar to RAID
mirroring and parity RAID, with automatic failover and recovery.
Compared with mergerfs, bcachefs is a full integrated filesystem with
built-in checksumming, snapshots, compression, and redundancy, while
mergerfs is a lightweight pooling layer over existing filesystems.
Bcachefs is still evolving and should be evaluated carefully for
production use.
The filesystem features comprehensive data protection including full
data and metadata checksumming, copy-on-write architecture that
eliminates write holes, and atomic transactions that ensure crash
safety. It also provides snapshots, transparent compression (LZ4,
gzip, Zstandard), and transparent encryption.
Bcachefs provides intelligent tiered storage where data can be
automatically placed across devices based on access patterns. Faster
devices (NVMe, SSD) can be designated as foreground or promotion
targets for hot data, while slower devices (HDD) serve as background
storage for cold data. This enables efficient use of heterogeneous
storage without manual configuration.
Like ZFS, bcachefs is most suitable for use cases where integrated
storage management, data redundancy, and reliability are important. It
differs from mergerfs in that it is a complete filesystem rather than
a pooling layer, providing built-in redundancy and automatic data
placement rather than relying on external tools or the properties of
underlying filesystems.
Bcachefs is under active development but has had a turbulent
relationship with the mainline kernel: it was merged into Linux 6.7
(January 2024) but subsequently removed in Linux 6.18 (late 2025)
following disagreements over development practices. It is now
distributed as an external DKMS module. It should be considered beta
quality and evaluated carefully before use in production systems.
* **Better than mergerfs for:** integrated CoW filesystem features,
checksumming, snapshots, compression, and built-in multi-device
redundancy.
* **Worse than mergerfs for:** conservative production setups prioritizing
maturity, simplicity, and broad long-term deployment stability.
## Btrfs Single Data Profile
* [https://btrfs.readthedocs.io](https://btrfs.readthedocs.io)
[Btrfs'](https://btrfs.readthedocs.io) `single` data profile is
similar to RAID0, spreading data across multiple devices but offering
no redundancy. Unlike mergerfs which pools existing filesystems at a
high level, Btrfs is a complete filesystem that manages storage
directly. If a single device fails in Btrfs single mode, you lose all
data. mergerfs takes a different approach: it pools filesystems as-is
without redundancy, so a device failure only affects data on that one
device, not the entire pool.
[Btrfs'](https://btrfs.readthedocs.io) `single` data profile is a
multi-device Btrfs mode that can spread data across drives without
adding redundancy.
Compared with mergerfs, Btrfs `single` is managed as one integrated
filesystem, while mergerfs pools separate existing filesystems. In
Btrfs `single`, losing one device can impact the whole filesystem;
with mergerfs, loss is typically limited to data on the failed disk.
* **Better than mergerfs for:** integrated multi-device management in one
filesystem namespace.
* **Worse than mergerfs for:** users who value per-disk failure isolation
and easy independent recovery.
## StableBit's DrivePool
* [https://stablebit.com](https://stablebit.com)
DrivePool is a commercial filesystem pooling technology for
Windows. If you are looking to use Windows then DrivePool is a good
option. Functionally the two projects work a bit
differently. DrivePool always writes to the filesystem with the most
free space and later rebalances. mergerfs does not currently offer
rebalance but chooses a branch at file/directory create
time. DrivePool's rebalancing can be done differently in any directory
and has file pattern matching to further customize the
behavior. mergerfs, not having rebalancing does not have these
features, but similar features are planned for mergerfs v3. DrivePool
has builtin file duplication which mergerfs does not natively support
(but can be done via an external program.)
DrivePool is a commercial storage pooling product for Windows that
combines multiple drives into a single volume and includes balancing
and optional duplication capabilities.
There are a lot of misc differences between the two projects but most
features in DrivePool can be replicated with external tools in
combination with mergerfs.
Compared with mergerfs, DrivePool emphasizes integrated balancing and
duplication features in a Windows-first product, while mergerfs
emphasizes Linux FUSE pooling with policy-based create placement and
external-tool composability.
* **Better than mergerfs for:** Windows users needing built-in
rebalancing and native file duplication features.
* **Worse than mergerfs for:** Linux-native deployments and users who
prefer open-source tooling and composable external utilities.
## Windows Storage Spaces
@ -311,20 +413,31 @@ solution with integrated data protection, Storage Spaces is a good
choice. For Linux users seeking lightweight pooling without redundancy
overhead, mergerfs is the better option.
* **Better than mergerfs for:** Windows environments needing integrated
pooling, resilience, and tiering in the platform stack.
* **Worse than mergerfs for:** Linux hosts and lightweight pooling where
minimal overhead and per-filesystem independence are preferred.
## Plan 9 binds
* [https://en.wikipedia.org/wiki/Plan_9_from_Bell_Labs#Union_directories_and_namespaces](https://en.wikipedia.org/wiki/Plan_9_from_Bell_Labs#Union_directories_and_namespaces)
Plan9 has the native ability to bind multiple paths/filesystems
together to create a setup similar to simplified union
filesystem. Such bind mounts choose files in a "first found" manner
based on the order they are configured similar to mergerfs' `ff`
policy. Similarly, when creating a file it will be created on the
first directory in the union.
Plan 9 has a native bind mechanism that can compose paths and
filesystems into a single namespace view, including union-like
directory behavior.
Compared with mergerfs, Plan 9 binds are an OS-native namespace feature
with simple ordered behavior, while mergerfs provides Linux-focused
filesystem pooling with a larger set of configurable branch policies.
Plan 9 isn't a widely used OS so this comparison is mostly academic.
* **Better than mergerfs for:** native Plan 9 environments where
namespace composition is a core OS feature.
* **Worse than mergerfs for:** Linux users due to ecosystem relevance and
practical deployment availability.
## SnapRAID pooling
@ -334,19 +447,28 @@ Plan 9 isn't a widely used OS so this comparison is mostly academic.
creates "a read-only virtual view of all the files in your array using
symbolic links."
As mentioned in the description this "view" is just the creation of
Compared with mergerfs, SnapRAID pooling is a symlink-generated,
read-only view refreshed by SnapRAID operations, while mergerfs is a
live mounted filesystem that supports normal read/write behavior.
As mentioned in the description, this "view" is just the creation of
the same directory layout with symlinks to all files. This means that
reads (and writes) to files are at native speeds but limited in that
it can not practically be used as a target for writing new files and
is only updated when `snapraid pool` is run. Note that some software
treat symlinks differently than regular files. For instance some
backup software will skip symlinks by default.
reads (and writes) to files are at native speeds, but it is not
practical to use as a target for writing new files, and it is only
updated when `snapraid pool` is run. Note that some software treats
symlinks differently than regular files. For instance, some backup
software will skip symlinks by default.
mergerfs has the feature [symlinkify](config/symlinkify.md) which
provides a similar behavior but is more flexible in that it is not
read-only. That said there can still be some software that won't like
that kind of setup.
* **Better than mergerfs for:** read-only, native-speed directory views
of SnapRAID-managed content.
* **Worse than mergerfs for:** general read/write pooled filesystem use
and always-live view updates.
## rclone union
@ -354,15 +476,24 @@ that kind of setup.
rclone's [union](https://rclone.org/union) backend allows you to
create a union of multiple rclone backends and was inspired by
[mergerfs](https://rclone.org/union/#behavior-policies). Given rclone
knows more about the underlying backend than mergerfs could it can be
more efficient than creating a similar union with `rclone mount` and
mergerfs.
[mergerfs](https://rclone.org/union/#behavior-policies). Because
rclone knows more about the underlying backends than mergerfs can, it
can be more efficient than creating a similar union with `rclone mount`
and mergerfs.
However, it is not uncommon to see users setup rclone mounts and
Compared with mergerfs, rclone union is backend-aware and optimized for
cloud/remote storage abstractions, while mergerfs is optimized for local
POSIX filesystem pooling and branch policies.
However, it is not uncommon to see users set up rclone mounts and
combine them with local or other remote filesystems using mergerfs
given the differing feature sets and focuses of the two projects.
* **Better than mergerfs for:** remote-backend-heavy workloads where
backend-aware behavior improves cloud storage handling.
* **Worse than mergerfs for:** local POSIX pooling scenarios where direct
filesystem semantics and broad local tool compatibility matter.
## distributed filesystems
@ -373,24 +504,20 @@ given the differing feature sets and focuses of the two projects.
* [MooseFS](https://moosefs.com)
* etc.
Distributed remote filesystems come in many forms. Some offering POSIX
filesystem compliance and some not. Some providing remote block
Distributed remote filesystems come in many forms. Some offer POSIX
filesystem compliance and some do not. Some provide remote block
devices or object stores on which a POSIX or POSIX-like filesystem is
built on top of. Some which are effectively distributed union
filesystems with duplication.
built. Some are effectively distributed union filesystems with
duplication.
These filesystems almost always require a significant amount of
compute to run well and are typically deployed on their own
hardware. Often in an orchestrator + worker node configuration across
numerous nodes. This limits their usefulness for casual and homelab
users. There could also be issues with network congestion and general
performance if using a single network and that network is slower than
the storage devices.
Compared with mergerfs, distributed filesystems focus on multi-node
storage, replication, and cluster behavior, while mergerfs focuses on
single-host local pooling of existing filesystems.
While possible to use a distributed filesystem to combine storage
devices (and typically to provide redundancy) it will require a more
complicated setup and more compute resources than mergerfs (while also
offering a different set of capabilities.)
* **Better than mergerfs for:** multi-node or geographically distributed
workloads with built-in replication and high availability.
* **Worse than mergerfs for:** single-host and homelab pooling where low
overhead and operational simplicity are key.
## nofs
@ -406,11 +533,20 @@ selection (mfs, ff, pfrd, epmfs, and others), read-only and no-create
branch modes, and recreates POSIX-like commands such as `ls`, `find`,
`which`, `cp`, `mv`, and `rm`.
Given its design nofs is not suited for general usage as 3rd party
Compared with mergerfs, nofs is command-driven rather than a mounted
filesystem, so mergerfs is the option for transparent system-wide
pooling through normal POSIX filesystem calls.
Given its design nofs is not suited for general usage as third-party
applications will not be able to take advantage of the unioning
behavior it offers. It is primarily for more simple situations where
something like mergerfs is unable to be used.
* **Better than mergerfs for:** constrained environments where a FUSE
mount is not possible but command-style union behavior is acceptable.
* **Worse than mergerfs for:** transparent system-wide filesystem pooling
usable by arbitrary third-party applications.
## policyfs
@ -423,23 +559,15 @@ distinguishing features are explicit path-pattern-based routing rules
for reads and writes, and an optional SQLite metadata index that can
serve `readdir` and `getattr` operations without spinning up HDDs.
Where mergerfs applies a single policy across all operations (with
per-function overrides), policyfs lets you write explicit routing
rules matched against path patterns, giving finer-grained control over
which physical storage a particular directory subtree reads from or
writes to. The SQLite metadata index is a notable addition for
HDD-heavy setups where keeping disks spun down is a priority (though
how effective it is will depend on access patterns); mergerfs does not
have an equivalent cache (though does leverage kernel caches when
enabled). policyfs also supports deferred physical operations,
recording delete and rename events for indexed paths and applying them
later.
Compared with mergerfs, policyfs emphasizes explicit path-based routing
and optional metadata indexing, while mergerfs emphasizes a broad,
policy-driven pooling model with longer operational history and wider
adoption.
policyfs is a young project (first released in late 2025) and should
be evaluated accordingly for production use. mergerfs has a longer
track record, a broader policy set, and wider community adoption. If
spin-down-friendly metadata serving or explicit per-path routing rules
are important to your workload, policyfs might be an option.
* **Better than mergerfs for:** explicit path-based routing and optional
metadata indexing to reduce HDD spin-ups.
* **Worse than mergerfs for:** users prioritizing project maturity,
adoption, and long-established operational behavior.
## Greyhole
@ -450,24 +578,16 @@ are important to your workload, policyfs might be an option.
Greyhole is an open-source storage pooling application that works via
Samba. Rather than implementing a FUSE filesystem, it hooks into Samba
VFS to intercept file operations and distribute files across multiple
drives. Its headline feature compared to mergerfs is per-share
redundancy: you can configure how many copies of each file Greyhole
keeps, with copies spread across different physical drives to protect
against drive failure.
drives, with optional per-share redundancy.
Because Greyhole operates through Samba, all file access must go
through Samba shares; it cannot be used as a general-purpose local
filesystem mount. mergerfs is a FUSE filesystem and works with any
application that uses normal POSIX filesystem calls. Greyhole is also
not well-suited for large numbers of small files or frequently-changing
files, and write performance is lower because files first land in a
temporary "landing zone" before being moved to the pool.
Compared with mergerfs, Greyhole is Samba-centric and includes
integrated share-level redundancy, while mergerfs is a general-purpose
Linux FUSE mount for local pooling via standard POSIX filesystem calls.
For users who need storage pooling with built-in per-file redundancy
and are already using Samba for file sharing, Greyhole provides both
in a single tool. For local filesystem pooling without Samba,
mergerfs is the more appropriate choice; redundancy can then be added
separately via SnapRAID or NonRAID.
* **Better than mergerfs for:** Samba-centric environments that need
built-in per-share redundancy in the same tool.
* **Worse than mergerfs for:** general-purpose local filesystem pooling
outside Samba and heavy small-file churn workloads.
## 9P
@ -475,10 +595,14 @@ separately via SnapRAID or NonRAID.
* [https://en.wikipedia.org/wiki/9P_(protocol)](https://en.wikipedia.org/wiki/9P_(protocol))
[9P](https://en.wikipedia.org/wiki/9P_(protocol)) is a filesystem
protocol from the Plan 9 operating system. While historically
important, it's not directly relevant for users looking to pool
filesystems. 9P is a network protocol, whereas mergerfs uses FUSE,
which is designed specifically for implementing filesystems in
userspace on Linux. FUSE has become the modern standard for this
purpose and better supports Linux filesystem features. For practical
filesystem pooling, mergerfs is the standard choice on Linux.
protocol from the Plan 9 operating system used to access files over a
network between systems.
Compared with mergerfs, 9P is a network filesystem protocol, while
mergerfs is a local Linux FUSE filesystem for pooling directories and
filesystems under one mount with branch policies.
* **Better than mergerfs for:** lightweight network filesystem protocol
use between compatible systems.
* **Worse than mergerfs for:** local Linux filesystem pooling because it
does not provide mergerfs-like branch policies or pooling semantics.