Btrfs/ZFS/bcachefs under workloads classic benchmarks skip
bartosz.fenski.pl
bartosz.fenski.pl
Having that said I'm desperately trying to get REAL hardware to run that benchmark. With some successes ;)
Few months ago I got Hetzner machine from Kent Overstreet and I was able to finish 3 runs before machine died... Results: https://bartosz.fenski.pl/modern-fs-benchmark/real-hw/
Currently I've got even more interesting machine with tons of disks and I'm running new set of benchmarks but it's really in its initial stage.
https://bartosz.fenski.pl/modern-fs-benchmark/sas-hdd/ 2nd run in progress... one run on REAL hardware takes much more time than on GH runner so it's slow.
But this new hardware has also so many disks that the plan is to try also more complex, tiered cache topologies. I'm working on it.
I'm happy to answer any other questions, sources of every piece of this benchmark are freely available and I'm not saying they are 100% correct. I'm open to improvements.
I've been saying it for months, but eventually I'm going to move the automated builds off the 48 core monster and we'll be able to use that for automated perf testing too. The machine we just got has spindles for EC perf testing, but the Hetzner monster has very high end enterprise ssdd.
Also, just got done with the Rust for Linux conference, still not home but here's slides that still need reformatting: https://evilpiepirate.org/~kent/Kangrejos-2026-bcachefs.pdf
1. Dual Ext4 + external 32GB journal X4 pcie SSD (the prior winner of benchmark surveys)
2. Bare F2FS after a trim and SSD vendor software cache flush operation (it should be slower, but knowing how much slower on identical hardware could be interesting.)
3. DRBD across a 48U 100Gbps host rack (single X4 pcie data drive per host, OS on primary)
4. CephFS across a 48U 100Gbps host rack (single X4 pcie data drive per host, OS on primary)
Best regards =3
5. a ZFS dRaid configuration. There could be very different characteristics there with it using slabs.
Speaking of slabs, MS ReFS of you feel adventurous!
* 4 HDDs (for example dm-raid has read balancing optimized specifically for HDDs)
* 5 HDDs (classical raid should see no improvement but btrfs and bcachefs should balance the load)
* 4 SSDs
* 3 HDDs + 1 SSD no tiering
* 2 HDDs + 2 SSD no tiering
* 1 drive 10x larger than others (since how bcachefs and btrfs allocators work)
* nocowLayered storage systems with a RAID layer that makes N disks look like one big disk generally don't have visibility into which blocks are free and which are allocated so they must "scrub" all the disks on initialization and repair even if only 1% is used.
Disregard, I just saw you have some RAID10 tests in there so three SSDs won't be enough.
Some remarks:
1. Why does the CoW button remove XFS from the list? Even https://github.com/fenio/modern-fs-benchmark/blob/main/scrip... mentions it has reflink enabled
2. If you have the time, adding XFS + mdraid + dm-integrity [1] (in bitmap mode) as a comparison point against ZFS RAID-Zx might be an interesting data point. That's what I run, personally.
3. Did you give some thoughts to the I/O scheduler choice? Might matter a lot in some cases.
[1] https://www.kernel.org/doc/html/latest/admin-guide/device-ma...
1. XFS reflink is enabled and its reflink/CoW-break measurements do run. The dashboard button currently means “native/full-CoW filesystem family”, not “supports reflink”, but that distinction is not clear from the label. I’ll rename it to “Native CoW” and add a separate reflink-capable filter that includes XFS.
2. The current integrity comparison is XFS on LVM/dm-raid10 with dm-integrity in its default journal mode. It is not mdraid and not bitmap mode, so your suggested stack would be a genuinely different and useful data point. An md RAID5/6 over per-member bitmap-mode dm-integrity comparison against RAID-Z1/Z2 makes sense, with the weaker post-crash bitmap semantics documented.
3. I did not pin or record the scheduler, which is a reproducibility gap. The dedicated SAS machine currently has mq-deadline active on all HDDs and SSDs. I’ll add queue/scheduler metadata to results before considering separate scheduler variants, since it can strongly affect the mixed and latency-sensitive phases.
But the reasons I choose filesystems are more about reliability, failure modes, surrounding tooling, and so on.
Btrfs fails in several critical areas:
1. No way to accurately find free space
2. catastrophic failure on write if a volume fills up, the probability of which is greater because of #1
3. repair tools usually do not recover a corrupted volume and in my testing are most likely to render as damaged volume completely unreadable, which makes #2 worse
Put these things together and I can never trust Btrfs again. In the 9 years since I encountered these, I see no effort to fix them, just fooling around witg unimportant side details like performance tweaks.
Fix the critical issues first then make it faster.
For that you have to dig into the methodology, look at the code, look at user reports, etc.
But you can get a pretty good approximation just from the philosophies and attitudes of the engineers and what they're talking about.
The talk I just gave at the Rust for Linux conference was all about that - how do we make the system debugable, the community aspect of how we respond to bug reports and talk to users, the prep work for the Rust conversion and formal verification and how we're approaching all that.
Reliability doesn't come out of nowhere, "all bugs are shallow with enough eyeballs" really doesn't apply to filesystems. You just have to plan for it, come up with a methodology, and do the work.
https://arstechnica.com/gadgets/2021/09/examining-btrfs-linu...
<- 5Y ago.
It's not materially better now. The devs are in denial about the problems because lots of big users are saying "works fine on my machine."
Sure, if you have lots of backups, if you have huge volumes on huge disks and they never fill up...
But it's the default in Fedora, Spiral Linux, Garuda Linux, siduction and others. Personal distros for people's own PCs and those are not well-supported enterprise kit.
Don't ever let it fill up!
SUSE takes snapshots before packages are installed. The package manager cannot tell if the disk will fill up as a result because on Btrfs the `df` command lies.
If its Btrfs root fills up and the OS writes to it, it 100% will self-destruct, and the `btrfs-repair` tool (`fsck` replacement) cannot fix drives and usually makes the corruption worse and renders the drive unreadable.
This is why in the earlier thread about swapfiles...
https://news.ycombinator.com/item?id=49618087
... I advised 2 commenters not to recommend keeping the OS and data in a single big volume. I got downvoted for it. I was rude. Well, I was, but I stand by my comments even though I'm sorry for my tone when I made them.
A pragmatic fix would trigger read-only failure mode at 90% usage of the mount as a whole, with all subvolumes.
And you know what, that's a great idea.
I have a loopback mount on an fallocate file with a btrfs filesystem inside of it at work. I'm using send/receive to make sure there is hope of recovery.
I wrote this several years ago.
https://www.linuxjournal.com/content/btrfs-centos-living-loo...
Here is my journey: https://forum.cgsecurity.org/phpBB3/viewtopic.php?p=39143
I switched to ZFS and never Bad a Problem again.
It would mount read-only, then? So you could copy your stuff off onto other media? Because if so, it's better than my experiences with Btrfs.
I think you are sugarcoating the shit show that was bcachefs's history of involvement in the linux kernel. I mean, do I need to mention that the person was subjected to a code of conduct enforcement action due to his long history of abuse and unprofessional behavior?
Most normal users and especially servers have no reason to run latest upstream kernels
I think if you're not using baremetal for such tests, it's likely that the results are simply not comparable at all? What if another tenant is also using the disk?
If it's something as simple as a KVM hypervisor that only runs 1 test VM at a time (with no other load from anything else other than the basic systemd daemons, ssh daemon etc running on the hypervisor), the results could be very close to bare metal.
I can see it being very time consuming and annoying to do repeated manual bare metal OS installs and new partitioning/filesystem creation for such a large variety of tests.
The author does also say that performance isn't really the main thing but rather, data integrity:
In this case, given that the author's own disclaimer (above) already disclaims the numeric readings, I'm not sure how it's possible to make any inference on "shapes and ratios" derived from the numeric readings.
Well don't do that then. There's lots of other options. Probably the simplest is a single bare metal install on a simple filesystem on one device. run the filesystems under test on other storage dedicated to testing.
You could also boot into a network install and use local storage exclusively for testing.
Also take a look at tests on real hardware. There are not many of them but there are some. I pointed to them in my first answer.
Real hardware is used in: https://bartosz.fenski.pl/modern-fs-benchmark/real-hw/ https://bartosz.fenski.pl/modern-fs-benchmark/sas-hdd/
But unfortunatelly it's much more limited number of actual runs. sas-hdd is still in progress so numbers for it should increase over time.
I'm saying ZFS on another OS.
I could drop bricks on and cord pull those all day and they would not lose data. Which is a small ask for a filesystem IMO.
I learned about VDO today. [0][1] I'd never heard of it before, but I'm using Gentoo Linux and both the dm-vdo kernel module and the 'vdo' software provided by [1] are distributed by Gentoo... so this isn't some weirdo Red Hat thing. It looks like you manage and used these just like any other thin-provisioned LVM volume [2], but -like I said- I've not used this before, and have only just skimmed the docs, so it's possible that I'm missing something important.
[0] <https://docs.redhat.com/en/documentation/red_hat_enterprise_...>
[1] <https://github.com/dm-vdo/vdo>
[2] <https://docs.redhat.com/en/documentation/red_hat_enterprise_...>
zfs is shunned. can't really get around that copyright issue. bcachefs just hit a setback, which i am hopeful will eventually be resolved.
But there's a LOT of FUD about it.
I guess my specific "bubble" simply didn't have that issue for a long time
Well except for Ubuntu, one of the most popular Linux distros supporting it.
One can say it is “dangerous” and they haven’t been sued “yet” but the deed is done already.
I am using Ubuntu!
> Why do you care if other distros consider it too risky given different business positions/models?
I don't know. Why do people comment on HN? :shrug:? It's fun and interesting to have a conversion?
I was mainly responding to a GP post that said "zfs is shunned. can't really get around that copyright issue". It didn't have a subject in there. As in "I can't really get around" or "anyone can't get around". And I read it as the second version and pointed that at least one distro did get around it.
Eh. Just wait until Oracle goes bankrupt due to their AI misinvestments & see where ZFS rights end up.
You can mix devices of different sizes and types on bcachefs. You can have foreground and background devices to balance performance and also different compression settings for foreground and background transactions.
You can set replicas=N to the individual file or directory on bcachefs. For example files you can just re-download or re-build. Likewise you can set a higher number of copies to important files.
Not everyone gets it though, that's for sure.
And, if you want to know if it's mature, I'd trust the user reports over the one liners :)
There's a NAS appliance called NASty and it shares publicly usage stats: https://nasty-telemetry.pages.dev/ Those numbers are NASty alone. There are more users on other systems.
Sometimes starting fresh with one coherent codebase and all features design baked in from the start might be better.
Edit: I think this makes for some good reading: https://lkml.org/lkml/2025/8/9/427
I'm afraid not all opinions are created equal, this isn't design by committee :)
I can’t remember anyone earning the level of hostile treatment he did and getting kicked out.
I don’t see how it would ever happen.
You don't seem to be following it very closely, as apparently you failed to come across the history of abuse and hostile behavior of bcachefs's maintainer.
If a professional and mature developer who understands a release cycle is willing to step up as a maintainer, yes.
Meanwhile, I no longer have to stress about whether I'll be about to get bugfixes out, actual users seem happy and my life is far better :)
No. It's very simple. 1) do QA and follow the release cycle, 2) act like a person when interacting with other developers.
If you fail to do both, you should not be surprised that people go through great lengths to not have to deal with you and all your nonsense.
The fact that after all this time you still feel compelled to gaslight everyone with this victim mentality shows that you still have a long way to go to, and a lot of soul searching to do.
I was spending a lot of my time just doing QA for the rest of the kernel.
You seem to just be painfully misinformed - or you're outright trolling.
There is and have been many promising and exciting FS to replace the old boring ones, but for storage you not only want to avoid technical issues but also maintainer(s) drama...
The community infighting has sucked, but that's a thing that matters primarily for maintainers.
I think most users just want something that works.
To answer the original question, most people who care about their filesystem at all care about its stability. Not just "does it work now" but also "will it work and improve over time". Infighting puts the future at risk.
But you might want to check out the bus factor on btrfs too; when a maintainer says "but we've saved Facebook billions and billions of dollars!", calls for the other filesystem maintainer to be ejected from the community, then quits to join Anthropic a month later - that's not a vote of confidence.
I'd be very happy if people could just stop bringing up drama and us factors. We put it behind us a year ago, but it seems not everyone got the memo.
Btrfs regulars:
- 1 from Meta
- 1 from Oracle
- 4 from SuSe
- 2 from WDC
bcachefs:
- Kent Overstreet
I doubt there'd be any real interest in a bastardized fork that only exists so the deep pocketed vendors can get away with code dump and run.
* https://btrfs.readthedocs.io/en/latest/btrfs-man5.html#man-b...
What have they been doing for the last decade(+)?
Western Digital picked up this work last year and is slowly getting through the new design without the write hole, I think we'll see working raid 5/6 in the next year or two.
I've used RAID-Z1/2/3 at a couple of jobs: yes IOps is suck-y, but if it's for backups of other systems, or the central logging server, or network monitoring (Suricata, Snort), sometimes your priority is cheap/bulk.
I'm currently in the HPC space, and Lustre is a thing here, and it has tiered storage via policies: you can (e.g.) put your recent/hot data on NVMe, but older/colder bits on spinning rust on ZFS.
That said I certainly hope that one day the technical advantage of bcachefs will be so overwhelming that maybe the decision to remove it will be overturned. And if big vendors make it their default FS the bus factor will disappear (even if unofficially you'd still be the sole maintainer, but no one cares about that in the enterprise world...)
True. But when you look at this, isn't the deeper problem that Linux remains such a monolith, and there's a stark difference between "included" and "not included" in the kernel. It's now over 35 years old. The fact that we can't have stable APIs and develop more out-of-tree drivers is not a strength, it's a weakness.
Even old Unix systems like SVR4 managed to have stable, public driver interfaces, despite being rather proprietary. FreeBSD manages to have drivers in its ports tree, with a stable API for a given major version. What makes Linux so special that it can't manage this?
I understand all of the arguments about why this has to be so. But... they might have made sense in the early days, but after 35 years it screams of immaturity. Plenty of other systems, including other open source systems, manage to do this, including having versioned interfaces so things aren't set in stone. Linux remains right at the extreme end of guaranteeing nothing. I've long thought this was unnecessary and counterproductive.
and, i believe i read somewhere that this is not an inability to stabilize the API but a conscious decision to not promise a stable API expressly because it creates the effect i described.
Jokes on you then: my favourite FS can't be "randomly removed from the kernel unexpectedly after some OS update" because it never made it in — ZFS. :)
So that leaves btrfs which always seemed complex and brittle to me (compared to zfs at least)
I get that real hardware costs (author mentions EUR 70 a month for a suitable server), but without at least a baseline snapshot comparison run between real hardware, both SSD and HDD, and the sparse file-backed loop devices, it's hard to take much away from this.
Sadly the AI apocalypse isn't making stuff like this easy to do as a hobby.
I do have a lot more pull requests to merge than I did before. I don't know if you want to count "Kent isn't reviewing PRs fast enough" as drama :)
also tricks to make it easy to convert a root FS to ZFS now that Ubuntu Server 24.04 added native root-on-zfs support: https://github.com/pirate/zfsify
In fact they delivered the erasure coding for parity raid back in march this year.
The thing is that as soon as you seriously give a chance to Bcachefs you see how good it is. I can only tell you that mixing different device tiers and having a per-file/directory replication setting is a god send specially in these times where storage costs more than gold.
I'm pretty sure Bcachefs is amazing and better than Btrfs. I also think Zfs is amazing and better than Btrfs. Even so, I still use Btrfs because I know it is guaranteed to always be present on any Linux without any effort on my part.
> That might be the best thing happened to the project since now development can happen at its own pace without the clicky bait influencers.
A better approach might have been to just paused mainline merging instead of forcing being kicked out?
Eg "Hey Linus, Bcachefs is still in early development and I need to merge changes in a pace that is not compatible with Linux development process. So I'm going to pause for a while now and once it reaches maintenance status I will focus on submitting patches in a healthy pace that you can digest".
As for BTRFS I think its also pretty good. Its just that I have the impression its development is guided by the needs of its sponsors and sadly for us META doesn't need RAID5.
A lot of things were tried, people did try to mediate.
The particularly galling thing though was when I finally started looking - post split - comparing bcachefs PRs to other subsystems and especially XFS - I was being more conservative with what I considered a critical bugfix.
There was never a clear statement on what the issue was. What you guys got in public was about as much as I got.
All I can say is - going fast when you're stabilizing and getting bugfixes out the door is what you can and should be doing when you've invested in test coverage, test automation, keeping the codebase clean and asserted, and building up a community that works well together on testing and shaking things out.
I genuinely do not know what they were thinking.
If you are referring to why bcachefs was removed from the Linux kernel, here's a discussion on bcachefs being removed from the Linux kernel.
They already know what was discussed.
(Good? Bad? Indifferent? I don't know and I don't have a dog in this race. I'm just here connecting the dots.)
Yes, that's why those remarks on how it's a mystery how bcachefs was pulled from the kernel are perplexing. To me they sound like gaslighting.
100%. My system is rock solid and the last thing I need is rolling the dice after every update on whether my system will boot. https://www.reddit.com/r/archlinux/comments/eywcp7/linux_551...
I'm impressed with bcachefs's accomplishments though, and if they ever reconcile with the kernel I'll surely give it a fair shake.
Actual distro support, and doing it right with people actually communicating with each other, has always been a priority for the project.
How large is large? I've deleted files with sizes of tens to hundreds of GBs and not seen that, and can probably whip up a test with a single-digit TB file if motivated.
Do you perhaps have 'discard=sync' in your mount options, or are using a kernel earlier than 6.2, which is the version -according to the docs- where async discard became the default?
for 1TB compressed (probably 5tb uncompressed) it is reproducable 100% reliably for me.
Here is some discussion: https://www.reddit.com/r/btrfs/comments/1mok440/filesystem_l...
I made a ~3TB btrfs FS and mounted it with 'force-compress', put 5TB of zeros on it (which compressed down to like 160GB), and did a delete along with some concurrent operations on that same FS. Based on what I saw, btrfs doesn't "block access for minutes" while a large delete is in progress, but access to the volume that has the delete in progress is dreadfully slow. I used vim to create a new file in the mountpoint and saw that write delays were between ten and twenty seconds. Really bad, but still functional. Not at all blocked.
Someone in that Reddit discussion that you linked to says that all btrfs filesystems hang during an extremely large delete. This is not what happens for me. The only btrfs FS made slow was the one that had the ongoing delete. I have four other btrfs filesystems mounted and they're all just fine, whether or not they're on the same physical disk that has the ongoing delete. space_cache is v2 on all of my btrfs filesystems.
On my system, it looks like an events_unbound kworker was eating 100% of a single CPU while the big delete was in progress. No other kernel threads seemed to be consistently occupied.
For fun, I re-ran the thing I document below when mounted without compression, and then with an uncompressable file when mounted with non-forced compression. I had to reduce the size of the file to 2TB for both scenarios, but omitting compression writes out like 10x the data to disk, so it still seems like a fair test.
I'm not going to the trouble to provide a transcript for those two runs, but both when mounted without compression enabled and when an uncompressable file was written to a compression-not-forced volume, I saw absolutely no delays in filesystem operations while I was deleting that 2TB file. FWIW, putting 2TB of /dev/zero on that compression-not-forced volume and deleting it behaved the same as it did on a 'force-compress' mount.
Whatever is causing the dreadful slowness is directly linked to transparent compression, rather than being something you get when you run btrfs in all configurations. "Why run btrfs if not for transparent compression?" you might ask. I would answer: "Snapshots and reflinks, and yes, I know that XFS has reflinks too.".
A lightly-edited terminal transcript follows for if you want to double-check my work up to the end of the 'force-compress' run.
# lvcreate --size 3T --name testlv --stripes=2 testvg
Using default stripesize 64.00 KiB.
Logical volume "testlv" created.
# mkfs.btrfs /dev/mapper/testvg-testlv
btrfs-progs v7.1
See https://btrfs.readthedocs.io for more information.
<extra crap removed>
# mount -o compress-force /dev/mapper/testvg-testlv /mnt/test/
# btrfs fi usage
Device size: 3.00TiB
Device allocated: 2.02GiB
Device unallocated: 3.00TiB
Device missing: 0.00B
Device slack: 0.00B
Used: 320.00KiB
Free (estimated): 3.00TiB (min: 1.50TiB)
<extra crap removed>
# dd if=/dev/zero of=/mnt/test/5TBFile bs=4MiB count=5TiB
1310720+0 records in
1310720+0 records out
5497558138880 bytes (5.5 TB, 5.0 TiB) copied, 2510.22 s, 2.2 GB/s
# /usr/bin/time --format='** fi sync %e' btrfs fi sync /mnt/test
** fi sync 0.00
# btrfs fi df /mnt/test/ | grep Data
Data, single: total=160.00GiB, used=160.00GiB
# /usr/bin/time --format='** totalTime %e' bash -c "
/usr/bin/time --format='** 20gbTime %e' bash -c 'dd if=/dev/zero of=/mnt/test/20GBFile bs=4MiB count=20GiB status=none; du -h /mnt/test/20GBFile; rm /mnt/test/20GBFile'&
/usr/bin/time --format='** 10gbTime %e' bash -c 'dd if=/dev/zero of=/mnt/test/10GBFile bs=4MiB count=10GiB status=none; du -h /mnt/test/10GBFile; rm /mnt/test/10GBFile'&
wait"
10G /mnt/test/10GBFile
** 10gbTime 2.42
20G /mnt/test/20GBFile
** 20gbTime 4.95
** totalTime 4.95
# date
Sun Sep 20 10:19:46 PM PDT 2026
# /usr/bin/time --format='** totalTime %e' bash -c "
/usr/bin/time --format='** 05tbTime %e' rm /mnt/test/5TBFile &
sleep 5 # It takes a few seconds for the delete to start making things slow when the file has been compressed. The operations on the 10GB file will complete in a normal amount of time if this sleep isn't present.
/usr/bin/time --format='** 20gbTime %e' bash -c 'dd if=/dev/zero of=/mnt/test/20GBFile bs=4MiB count=20GiB status=none; du -h /mnt/test/20GBFile; rm /mnt/test/20GBFile'&
/usr/bin/time --format='** 10gbTime %e' bash -c 'dd if=/dev/zero of=/mnt/test/10GBFile bs=4MiB count=10GiB status=none; du -h /mnt/test/10GBFile; rm /mnt/test/10GBFile'&
wait"
10G /mnt/test/10GBFile
** 10gbTime 183.64
20G /mnt/test/20GBFile
** 20gbTime 259.47
** 05tbTime 1026.21
** totalTime 1026.22
# date ; /usr/bin/time --format='** 20gbTime2 %e' bash -c 'dd if=/dev/zero of=/mnt/test/20GBFile bs=4MiB count=20GiB status=none; du -h /mnt/test/20GBFile; rm /mnt/test/20GBFile' ; date
Sun Sep 20 10:36:52 PM PDT 2026
20G /mnt/test/20GBFile
** 20gbTime2 23.04
Sun Sep 20 10:37:15 PM PDT 2026I have 'discard=async'
I used to think that about ReiserFS, too. It was in the mainline kernel, development was snappy, and it solved some performance problems. I used it all over the place.
Things then subsequently... changed. :-/
And I did move away from it.
But I had once expected ReiserFS to be permanent, especially since this Hans Raiser dude who was driving the ship seemed to be sharp AF, and this seemed doubly-true when his filesystem got mainlined.
But it was not permanent. It did not last forever.
This idea of permanence, or rather the lack of it, was the whole of the point that I was responding to and also trying to impress upon.
At the end of the day: We do not know the future. Things can change.
I've outlived this one high-performance Linux filesystem in my life that I was using. This does not in any way mean that I will be the last to outlive other Linux filesystems.
Permanence is not guaranteed.
Except RHEL. They don’t include it in their kernels.
Alma Linux started including it again though.
It can never be easy.
Instead it got kicked out because Kent constantly ignored the kernel's contribution rules and is unlikely it will ever be accepted back into the kernel.
And it went in when it did because Redhat was pushing for it and claiming to be supportive - but that never materialized. They wanted to get something for free without investing, or putting in the absolute bare minimum.
A _lot_ of people were saying publicly and privately "dear god yes we need something better than btrfs" - but no one from the existing kernel community was interested in stepping up.
Community's still growing, though. A lot of people have gotten active in making sure bcachefs actually works well for people end to end, and there's a hell of a lot more to shipping a filesystem than just writing kernel code.
The FS was marked experimental, so there is no urgency in fixing bugs or providing features in a certain cycle. Everyone using it knows what they got themselves into. You can still provide the DKMS module for faster fixes and features for anyone who wants to use BCacheFS more seriously for the time that the upstreaming process takes, but eventually it would have all been on mainline.
Asahi is taking a similar approach where they have their downstream kernel and push things upstream once they are mature.
That means the upstream kernel is not useful for running on that hardware now, but things are moving there eventually.
All this has been discussed to death, we don't need people armchair quarterbacking a year later. It's over, it's time to move on.
Thanks for pointing it.
How is root support for bcachefs? This is the one thing I really miss with ZFS today, its just too much work. I do love my FreeBSD systems with a root ZFS though.
I say max, as ZFS stores multiples of the physical block size up to the record size. So even with 128k record size, if your file fits in two physical blocks, it will write two physical blocks of data.
The `ashift` parameter[2] controls the physical block size, typically 512 bytes or 4k for HDDs. Though higher can be useful on SSDs (but less tested, have seen some bug reports with >4k block sizes).
Typically you'd only want to use such small records if you are tuning it for a specific workload, like if you have a database that writes 8k pages.
Compression acts on records, so by limiting the record size, you limit the effectiveness of the compression: 1.2 and 1.8 blocks worth of data both gets written as 2 physical blocks.
On the upside, record size is a dataset property, so you can have many different datasets with different record sizes on the same pool.
[1]: https://github.com/fenio/modern-fs-benchmark/blob/599ec72fe3...
[2]: https://openzfs.github.io/openzfs-docs/Performance%20and%20T...
Also hope Proxmox decides to incorporate it somewhat soon in PVE.
I'm not sure that nuking 2G of the underlying block device is a recoverable error on any filesystem that I'm aware of? Can you confirm if any ofthe filesystems really came out of the other side in a usable state after scrubbing?
-----
> Trivial-op p99, idle (ms) # A trivial operation — one 4k write + fsync every 200ms (like a shell appending history or an editor updating its swap file) — run alone for 10s. p99 of the fsync completion
In fact, if it's OK for me to ask, are any of the metrics tht you used standard industry metrics? It looks like several of the tests are bypassing the kernel's page cache? -- which I worry may fall into the trap of "I modified the system to be unrepresentative of reality and then tested it".
----
> kernel 7.0.0-1012-azure
Can you confirm if you tested on a bare metal machine? were you the only tenant?
ZFS and btrfs were designed from the start to handle this, by using checksums on every piece of (meta)data and redundancy to return the same data as was stored to the kernel, and rewrite the bad data.
I've tested my own machines running ZFS by random writes out of band from the filesystem/kernel and it has always found and fixed them.
Speaking about standard industry metrics.
fio is an established tool, and throughput, IOPS, fsync latency, and percentiles are standard concepts. However, the exact job recipes and the composite “Overall Core” score are project-specific.
The trivial-operation test is also custom: one 4 KiB write plus fsync every 200 ms. The idle window contains at most about 50 operations, so its p99 is effectively the slowest sample’s fio histogram bucket, not a statistically stable population percentile. It should be treated as a small-write durability-latency probe, not a universal application metric.
But after all all tests are in the repository. If they need tweaks, changes I'm open to do so... I started from scratch and did whatever came to my mind. Some tests are added after my initial link here which went mostly unnoticed several weeks ago but I got some requests for more tests which I implemented.
But to sum up. I want this test to be useful so feel free to open PRs with improvements. It's not like I've got some agenda. In fact I wrote here and there on the page that I'm counting on communities of various filesystems to provide improvements, changes etc to make their filesystem shining.
This is personal project made when I realized that multiple-devices benchmarks were almost completely absent. Since I had not access to real hardware I decided to make at least initially everything based on GH runner with all the limitations that came with this approach. I tried to limit these limitations as far as I could. But feel free to submit bugreports, PRs, propositions for improvements.
But if it helps, just some "initial gut feel observations" from me:
* It's definitely not possible to find issue with the the _sheer amount_ of results, but there's just far too much for a human to absorb, all presented at once
* Overall text size is quite small, and difficult to read
* The page doesn't make a strong statement of _what_ is under test: the first words are: "modern-fs-benchmark Multi-device CoW filesystems under workloads classic benchmarks skip" -- which defines the webpage in terms of what it is _not_, without stating what benchmarks are actually present.
* The first line of teh page contains run statistics that probably eithre want to b at the bottom, or just don't need to be in the webpage at all: "latest run 2026-09-18 18:50:45 UTC, kernel 7.0.0-1012-azure, 593 runs recorded · 145 trend points shown"
* A significant proportion of the free text is caveats. There's nothing wrong with being transparent about limitations, but they may be a sign that there might be alternative ways to present the data, or that the data may be flawed (depending on the caveat)
* Theres several categories that I think have been invented for the purpose of collation, but I don't think are defined on the page. I think "Overall Core" and "Core I/O" aren't explained, which means by definition it's impossible for a reader to understand the score table.
* And as we're all aware right now, current Claude models are currently struggling to write coherent English. There's several incoherent sentences on the page. It's a Claude issue.
Sometimes there are ways to make things easier without dumbing them down, but way too many people conflate the two; I get nervous when non engineers say "I've studied this, it should be easy".
At some point I'd like to get our own automated pts runs going, since Michael is not consistent with what hardware he tests on and he hasn't been consistent with getting them out.