11 ms·

So the two things I don't see are a comparison with BTRFS in terms of speed, and stability. I know ZFS is cool, but while we were waiting for it BTRFS got prett

by dkhenry 14y ago

So the two things I don't see are a comparison with BTRFS in terms of speed, and stability. I know ZFS is cool, but while we were waiting for it BTRFS got pretty darn good. I know ZFS has some additional features that BRTFS doesn't have, but If I have to use a unstable or slow filesystem to get those features when I could use by perfectly stable and fast filesystem while I wait for a few nice features I am going to say ZFS has missed its oppertuinity

wazoox 14y ago
People love to harp about the benefits and wonders of ZFS, but they often forgot to mention its limitations: most features have an impact on performance, memory consumption is ginormous, IO performance is generally not great (lower than with "simpler" FS), and managing your pools isn't that easy once you get serious about it.
- johnbellone 14y ago
  Not to mention the whole lack of an upstream. But all that aside ZFS was great when it was first introduced a few years back, especially with Sun backing OpenSolaris, but we've been jerked around for a few years. A ran a few ZFS pools on Linux a few builds ago and it was quite stable then. For something like a home storage system where I do not care about all the shortfalls vs. BTRFS I would reach to ZFS just because its what I know. For something that needs to be enterprise I'd hire someone that knows much more about it than me :).
  - dmpk2k 14y ago
    we've been jerked around for a few years Oracle's ZFS (and Solaris) fork is a deadish end. Most of the interesting action the past couple years has been happening among a collection of illumos (including critical ZFS members formerly from Sun), FreeBSD, OSX and even Linux hackers. I don't follow their activities closely, except to observe that they've been anything but idle. A lot of work has happened in the meantime, and continues. Hopefully one shows up to comment about it.
    - newman314 14y ago
      List of different versions of ZFS https://en.wikipedia.org/wiki/ZFS#Release_history https://en.wikipedia.org/wiki/ZFS#Release_history
    - natermer 14y ago
      Oracle's ZFS is now different then BSD/Linux ZFS. They have become incompatible file systems because Oracle has picked up their ball and taken it back home. For OSS ZFS the only upstream is 'collection of Illlumos'
    - johnbellone 14y ago
      I was running OpenIndiana for awhile at home before I just got tired of trying to hack together a bunch of patches. The problem with having a project that is pseudo OSS is that you can never actually use it seriously, its never reliable and for the most part its going to end in slow quiet death. In my opinion there are a were only a handful of reasons to use Solaris: Oracle, SPARC, ZFS and dtrace. The only reason to use it nowadays is because you can't scale x86 and need big iron for Oracle et al.
      - dmpk2k 14y ago
        Unrelated to the topic at hand, but I've been happy with SmartOS. It's firmly and solely an OSS server OS, it's easy to use, many of the uglier Solaris legacy bits have been sanded off, and it has a decent package system (pkgsrc). My earlier experiences with Solaris and OpenSolaris were mixed. SmartOS has been uniformly positive. Perhaps give it a try, although be aware that it (like everything illumos) is picky about the hardware.
- trotsky 14y ago
  Performance seems ok to me. 6x500gb 7.2k sata raid10 + 96gb ssd, 8gb ram. Run began: Fri Mar 29 08:32:24 2013 Include fsync in write timing Include close in write timing Command line used: iozone -ec -r 32 -s 16380m -l 2 -i 0 -i 1 -i 8 Throughput test with 2 processes Each process writes a 16773120 Kbyte file in 32 Kbyte records Parent sees throughput for 2 initial writers = 996073.19 KB/sec Avg throughput per process = 499898.70 KB/sec Parent sees throughput for 2 rewriters = 216128.36 KB/sec Avg throughput per process = 108075.14 KB/sec Parent sees throughput for 2 readers = 1232245.62 KB/sec Avg throughput per process = 616353.25 KB/sec Parent sees throughput for 2 re-readers = 1240045.15 KB/sec Avg throughput per process = 620251.62 KB/sec
  - dekhn 14y ago
    that's really not so impressive. you're just dumping extents down on an empty filesystem and re-reading them. you're really just measuring the fact that the FS isn't particularly getting in the way of using your hardware. Note also that rewriting drops to 108M/sec, which is pretty unimpressive.
    - trotsky 14y ago
      216M/sec - rewrite is suffering from a striding mismatch there with the COW, bs is 32k while stripe is 128k. Consumer uses 1M. I'm wondering which volume manager and fs under linux you'd expect better performance from in this configuration?
- craigyk 14y ago
  Yeah... what are you talking about? Enormous memory consumption = deduplication? Otherwise it doesn't seem enormous too me. Sure ZFS made a lot of departures from the standard toolset with all its own stuff... but that stuff is better once you learn it. I can easily get 600MB+/sec reads and writes on my pools with are pretty commodity stuff. One of my pools is 95% full, which is high, and I just tested it and got over ~100MB/sec write ~150MB read.... good enough for me. (Oh, and I just checked, it was scrubbing) Here are my little ZFS peeves: 1. you can't grow raidZ vdevs. Reduces flexibility. 2. I wish there was a version of scrub that rebalanced data across vdevs. By comparison BTRFS is nice, but nowhere close to ZFS. Last I saw a BTRFS scrub consists of cat'ing every file to null... which doesn't even cover the FS and metadata.
- glabifrons 14y ago
  "memory consumption is ginormous" Yes, it will use whatever memory you allow it to, but this is purely dynamic just as caching is in Linux. If you're not talking about the ARC, but about deduplication, of course it uses more memory - how would it not? "IO performance is generally not great (lower than with "simpler" FS)" This is sounding troll-ish, as that statement equates to: A filesystem that checksums all data and metadata and performs copy-on-write to protect the integrity of the on-disk state at all times is slower than a filesystem that does neither. Well, of course. But do those "simpler FS" have the ability to massively negate that effect by use of an SSD for the ZIL and L2ARC? There have been many articles showing higher throughput with large, slow 5400RPM drives combined with an SSD ZIL & L2ARC massively outperforming much faster enterprise drives. "managing your pools isn't that easy once you get serious about it" I'm fairly stunned by this statement, as I've yet to see an easier, more elegant solution for such a task. Before ZFS, I liked VXVM with VXFS, but I now consider it obsolete. Linux's LVM is downright painful in comparison. I've yet to play with btrfs, so I'll bite my tongue on what I've read so far on it. The deep integration of volume and pool management directly with the filesystem, essentially making them one and the same, is simply beautiful. Having these things separate (md, LVM, fs) after years of using ZFS seems so archaic and awkward to me. Disclosure: 100% of my ZFS experience has been on Solaris (10-11.1) and OpenSolaris/Nevada. I've not tried it on Linux, yet.
kev009 14y ago
Btrfs is actually a sordid state of affairs. It's been "two years out" for five years, which seems very disingenuous looking back. Groundbreaking FS development follows a pretty regular formula. 10 years seems to be the magic number. Major functionality (i.e RAID 5) is still just landing and has glaring issues. It's just now a couple years out. Meanwhile, ZFS has over 10 years of history, stably implements most of Btrfs _planned_ features, and has battle tested deployments. Granted, the SPL for Linux adds variables, but there are some big users of this particular project. So I approach Btrfs with the exact opposite mindset. It's guilty until innocent despite some FUD from the Linux camp early on that has settled down a bit since Oracle now has it's hands on both.
- simplexion 14y ago
  i.e. == that is. e.g. = example. RAID5 is not major functionality, it should no longer exist.
  - Dylan16807 14y ago
    Can you explain why you feel that way? ZFS has intelligently managed RAID that keeps the performance penalty minimal. I'd rather have multiple-redundancy parity than 2x mirroring, and it takes less space.
    - gtirloni 14y ago
      ZFS' raid-z is nothing but pain. Does it have standard raid-5 now?
      - simplexion 14y ago
        There is RAID-Z2 which allows for 2 discs to fail and RAID-Z3 which allows for 3. Problem is rebuild time is still slow. I wouldn't recommend to use RAID-Z1.
- dkhenry 14y ago
  This is why people think BTRFS isn't any good, because someone waltses up and says yeah well it doens't have XXX feature which only _one_ other filesystem has so its clearly lacking and immature. The truth it it works great its sufficently fast and has been stable for years. Yes it doesn't have Raid5 or Raid Z if you would like and I can't import and export pools the way I can in ZFS, but honestly if you jumping on a FS that has been marked stable for all of two days on Linux and your reasoning is you can't do software raid on the other one then best of luck to you.
antonios 14y ago
Let me understand: You claim that ZFS is the unstable or slow filesystem, and that BTRFS, which is not even yet marked as stable by its developers is the perfectly stable and fast one? That seems an opposite way of thinking.
- icebraining 14y ago
  Not ZFS, but ZFS for Linux.
brokenparser 14y ago
I tried resizing btrfs recently, my file came out truncated :(
__david__ 14y ago
> I know ZFS is cool, but while we were waiting for it BTRFS got pretty darn good. My own anecdotal evidence is that btrfs is not ready for prime time. I ran a couple btrfs file systems from kernel about 2.6.30 to 3.4 and finally determined that my system was so slow because of btrfs. I switched to xfs and an operation that used to take weeks (literally) now takes hours. Perhaps it's gotten significantly better in the last few revisions but I'd use caution if you want to use btrfs and still have good performance.
65a 14y ago
I'm getting +20k IOPs and nearly 1GB/sec read/write out of a mirror, something like 40K 1.5GB/sec out of a mirror+stripe on ZFS under FreeBSD. Disks are either 2x256GB SSD or 4x128GB SSD. Meanwhile BTRFS was so slow on random IO I had to toss it out in favor of the less reliable JFS or EXT4 on Gentoo. Don't use RAIDZ if you want performance with ZFS however, and you need to set it up twice before you get how to performance tune it.

by dkhenry 14y ago

wazoox 14y ago

People love to harp about the benefits and wonders of ZFS, but they often forgot to mention its limitations: most features have an impact on performance, memory consumption is ginormous, IO performance is generally not great (lower than with "simpler" FS), and managing your pools isn't that easy once you get serious about it.

johnbellone 14y ago
Not to mention the whole lack of an upstream. But all that aside ZFS was great when it was first introduced a few years back, especially with Sun backing OpenSolaris, but we've been jerked around for a few years. A ran a few ZFS pools on Linux a few builds ago and it was quite stable then. For something like a home storage system where I do not care about all the shortfalls vs. BTRFS I would reach to ZFS just because its what I know. For something that needs to be enterprise I'd hire someone that knows much more about it than me :).
- dmpk2k 14y ago
  we've been jerked around for a few years Oracle's ZFS (and Solaris) fork is a deadish end. Most of the interesting action the past couple years has been happening among a collection of illumos (including critical ZFS members formerly from Sun), FreeBSD, OSX and even Linux hackers. I don't follow their activities closely, except to observe that they've been anything but idle. A lot of work has happened in the meantime, and continues. Hopefully one shows up to comment about it.
  - newman314 14y ago
    List of different versions of ZFS https://en.wikipedia.org/wiki/ZFS#Release_history https://en.wikipedia.org/wiki/ZFS#Release_history
  - natermer 14y ago
    Oracle's ZFS is now different then BSD/Linux ZFS. They have become incompatible file systems because Oracle has picked up their ball and taken it back home. For OSS ZFS the only upstream is 'collection of Illlumos'
  - johnbellone 14y ago
    I was running OpenIndiana for awhile at home before I just got tired of trying to hack together a bunch of patches. The problem with having a project that is pseudo OSS is that you can never actually use it seriously, its never reliable and for the most part its going to end in slow quiet death. In my opinion there are a were only a handful of reasons to use Solaris: Oracle, SPARC, ZFS and dtrace. The only reason to use it nowadays is because you can't scale x86 and need big iron for Oracle et al.
    - dmpk2k 14y ago
      Unrelated to the topic at hand, but I've been happy with SmartOS. It's firmly and solely an OSS server OS, it's easy to use, many of the uglier Solaris legacy bits have been sanded off, and it has a decent package system (pkgsrc). My earlier experiences with Solaris and OpenSolaris were mixed. SmartOS has been uniformly positive. Perhaps give it a try, although be aware that it (like everything illumos) is picky about the hardware.
trotsky 14y ago
Performance seems ok to me. 6x500gb 7.2k sata raid10 + 96gb ssd, 8gb ram. Run began: Fri Mar 29 08:32:24 2013 Include fsync in write timing Include close in write timing Command line used: iozone -ec -r 32 -s 16380m -l 2 -i 0 -i 1 -i 8 Throughput test with 2 processes Each process writes a 16773120 Kbyte file in 32 Kbyte records Parent sees throughput for 2 initial writers = 996073.19 KB/sec Avg throughput per process = 499898.70 KB/sec Parent sees throughput for 2 rewriters = 216128.36 KB/sec Avg throughput per process = 108075.14 KB/sec Parent sees throughput for 2 readers = 1232245.62 KB/sec Avg throughput per process = 616353.25 KB/sec Parent sees throughput for 2 re-readers = 1240045.15 KB/sec Avg throughput per process = 620251.62 KB/sec
- dekhn 14y ago
  that's really not so impressive. you're just dumping extents down on an empty filesystem and re-reading them. you're really just measuring the fact that the FS isn't particularly getting in the way of using your hardware. Note also that rewriting drops to 108M/sec, which is pretty unimpressive.
  - trotsky 14y ago
    216M/sec - rewrite is suffering from a striding mismatch there with the COW, bs is 32k while stripe is 128k. Consumer uses 1M. I'm wondering which volume manager and fs under linux you'd expect better performance from in this configuration?
craigyk 14y ago
Yeah... what are you talking about? Enormous memory consumption = deduplication? Otherwise it doesn't seem enormous too me. Sure ZFS made a lot of departures from the standard toolset with all its own stuff... but that stuff is better once you learn it. I can easily get 600MB+/sec reads and writes on my pools with are pretty commodity stuff. One of my pools is 95% full, which is high, and I just tested it and got over ~100MB/sec write ~150MB read.... good enough for me. (Oh, and I just checked, it was scrubbing) Here are my little ZFS peeves: 1. you can't grow raidZ vdevs. Reduces flexibility. 2. I wish there was a version of scrub that rebalanced data across vdevs. By comparison BTRFS is nice, but nowhere close to ZFS. Last I saw a BTRFS scrub consists of cat'ing every file to null... which doesn't even cover the FS and metadata.
glabifrons 14y ago
"memory consumption is ginormous" Yes, it will use whatever memory you allow it to, but this is purely dynamic just as caching is in Linux. If you're not talking about the ARC, but about deduplication, of course it uses more memory - how would it not? "IO performance is generally not great (lower than with "simpler" FS)" This is sounding troll-ish, as that statement equates to: A filesystem that checksums all data and metadata and performs copy-on-write to protect the integrity of the on-disk state at all times is slower than a filesystem that does neither. Well, of course. But do those "simpler FS" have the ability to massively negate that effect by use of an SSD for the ZIL and L2ARC? There have been many articles showing higher throughput with large, slow 5400RPM drives combined with an SSD ZIL & L2ARC massively outperforming much faster enterprise drives. "managing your pools isn't that easy once you get serious about it" I'm fairly stunned by this statement, as I've yet to see an easier, more elegant solution for such a task. Before ZFS, I liked VXVM with VXFS, but I now consider it obsolete. Linux's LVM is downright painful in comparison. I've yet to play with btrfs, so I'll bite my tongue on what I've read so far on it. The deep integration of volume and pool management directly with the filesystem, essentially making them one and the same, is simply beautiful. Having these things separate (md, LVM, fs) after years of using ZFS seems so archaic and awkward to me. Disclosure: 100% of my ZFS experience has been on Solaris (10-11.1) and OpenSolaris/Nevada. I've not tried it on Linux, yet.

kev009 14y ago

Btrfs is actually a sordid state of affairs. It's been "two years out" for five years, which seems very disingenuous looking back. Groundbreaking FS development follows a pretty regular formula. 10 years seems to be the magic number. Major functionality (i.e RAID 5) is still just landing and has glaring issues. It's just now a couple years out. Meanwhile, ZFS has over 10 years of history, stably implements most of Btrfs _planned_ features, and has battle tested deployments. Granted, the SPL for Linux adds variables, but there are some big users of this particular project. So I approach Btrfs with the exact opposite mindset. It's guilty until innocent despite some FUD from the Linux camp early on that has settled down a bit since Oracle now has it's hands on both.

simplexion 14y ago
i.e. == that is. e.g. = example. RAID5 is not major functionality, it should no longer exist.
- Dylan16807 14y ago
  Can you explain why you feel that way? ZFS has intelligently managed RAID that keeps the performance penalty minimal. I'd rather have multiple-redundancy parity than 2x mirroring, and it takes less space.
  - gtirloni 14y ago
    ZFS' raid-z is nothing but pain. Does it have standard raid-5 now?
    - simplexion 14y ago
      There is RAID-Z2 which allows for 2 discs to fail and RAID-Z3 which allows for 3. Problem is rebuild time is still slow. I wouldn't recommend to use RAID-Z1.
dkhenry 14y ago
This is why people think BTRFS isn't any good, because someone waltses up and says yeah well it doens't have XXX feature which only _one_ other filesystem has so its clearly lacking and immature. The truth it it works great its sufficently fast and has been stable for years. Yes it doesn't have Raid5 or Raid Z if you would like and I can't import and export pools the way I can in ZFS, but honestly if you jumping on a FS that has been marked stable for all of two days on Linux and your reasoning is you can't do software raid on the other one then best of luck to you.

antonios 14y ago

Let me understand: You claim that ZFS is the unstable or slow filesystem, and that BTRFS, which is not even yet marked as stable by its developers is the perfectly stable and fast one? That seems an opposite way of thinking.

icebraining 14y ago
Not ZFS, but ZFS for Linux.

brokenparser 14y ago

I tried resizing btrfs recently, my file came out truncated :(

__david__ 14y ago

> I know ZFS is cool, but while we were waiting for it BTRFS got pretty darn good. My own anecdotal evidence is that btrfs is not ready for prime time. I ran a couple btrfs file systems from kernel about 2.6.30 to 3.4 and finally determined that my system was so slow because of btrfs. I switched to xfs and an operation that used to take weeks (literally) now takes hours. Perhaps it's gotten significantly better in the last few revisions but I'd use caution if you want to use btrfs and still have good performance.

65a 14y ago

I'm getting +20k IOPs and nearly 1GB/sec read/write out of a mirror, something like 40K 1.5GB/sec out of a mirror+stripe on ZFS under FreeBSD. Disks are either 2x256GB SSD or 4x128GB SSD. Meanwhile BTRFS was so slow on random IO I had to toss it out in favor of the less reliable JFS or EXT4 on Gentoo. Don't use RAIDZ if you want performance with ZFS however, and you need to set it up twice before you get how to performance tune it.