this post was submitted on 18 Mar 2026
93 points (98.9% liked)

Linux

12928 readers
550 users here now

A community for everything relating to the GNU/Linux operating system (except the memes!)

Also, check out:

Original icon base courtesy of lewing@isc.tamu.edu and The GIMP

founded 2 years ago
MODERATORS
 

Last week I provided a look at the EXT4 and XFS performance from Linux 6.12 LTS through Linux 7.0 in its current development form. As mentioned in that article and as requested by many Phoronix readers, benchmarks have since wrapped up looking at how the Btrfs copy-on-write file-system performance has evolved since that late 2024 period and all major Linux kernel releases past that Long Term Support version.

top 16 comments
sorted by: hot top controversial new old
[–] KindaABigDyl@programming.dev 30 points 5 days ago* (last edited 5 days ago) (3 children)

Maybe this is wrong, but my understanding is BTRFS is generally slower than EXT4, and that's okay. It's not going for speed

Where it shines is not in its speed but in its versatility offering compression, rollback, subvol, etc

For example, for applications, you do a lot of writes/reads to Documents or load resources like for games, so use EXT4 for /home or for a dedicated /games partition

For your system, it could be broken via config tweaks or updates, so use BTRFS to have the rollback options

[–] morto@piefed.social 16 points 5 days ago (1 children)

I'm still amazed from discovering filesystem-level compression.

[–] vividspecter@aussie.zone 7 points 5 days ago (1 children)

cp --reflink is really cool as well, even if it's just something you expect from a CoW filesystem. Being able to near instantly duplicate a file or folder and work on its copy (with only the changes taking up space) is very useful.

[–] morto@piefed.social 9 points 5 days ago

That's great too!

y'all got any more ot that btrfs tricks meme

[–] ExtremeUnicorn@feddit.org 4 points 5 days ago

Also, it has self-healing RAID capabilities.

[–] Blue_Morpho@lemmy.world 5 points 5 days ago (1 children)

It's not that it's slower than ext4 but that btrfs itself has gotten slower.

[–] KindaABigDyl@programming.dev 1 points 4 days ago

Right but what I'm saying is speed wasn't really the reason to use it in the first place

[–] lnxtx@sopuli.xyz 7 points 5 days ago

Time to git bisect ಠ_ಠ

[–] Jankatarch@lemmy.world 3 points 5 days ago* (last edited 5 days ago) (1 children)

Didn't the lead contributor get hyper-psychosis or was that a different filesystem?

[–] cm0002@suppo.fi 47 points 5 days ago (2 children)

ai psychosis or something, but it was Kent Overstreet of Bcachefs

Btrfs is different

[–] qjkxbmwvz@startrek.website 27 points 5 days ago (2 children)

...and regular old murder for ReiserFS.

Not sure what it is about filesystem maintainers...

[–] cm0002@suppo.fi 17 points 5 days ago

WHAT

note to self, do not develop OS's or filesystems lmao

[–] Jankatarch@lemmy.world 9 points 5 days ago* (last edited 5 days ago)

Ah thanks I mixed up the projects.

[–] csolisr@hub.azkware.net 2 points 5 days ago (1 children)

Ah bummer... I'm currently using XFS on my server because BTRFS's performance is less than stellar, although I'd love to use it to have native compression

[–] poinck@lemmy.world 1 points 4 days ago* (last edited 4 days ago)

I have use cases for btrfs, xfs and zfs. Somehow ext4 feels legacy or for small systems like Raspberries or when the cloud-image provided is already ext4.

I use BTRFS for personal PCs because of the subvolume feature (since one year or so), ZFS for backup/archive when I need raid and encryption capability without hardware raid and for proxmox. XFS is for large storage servers where hardware raid is already established or very special cases when a lot of inodes are needed.