Announcement

Collapse
No announcement yet.

Oracle Now Supports Btrfs RAID5/6 On Their Unbreakable Enterprise Kernel

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by starshipeleven View Post
    a filesystem that has CoW and checksumming only on metadata cannot be compared with one that is CoW on everything (btrfs/zfs) AND can software-raid on its own.
    Why can't you compare them again? One has full CoW, the other hasn't. There, they are compared. Nobody said APFS matches BTRFS or ZFS feature by feature.

    I still don't see the point in CoW and checksumming only metadata. Isn't a journal enough?
    Just because you don't see the point doesn't mean there isn't one. The keyword here is speed. Journaling has overhead.

    CoW and checksumming of metadata are comparable to NTFS's journaling and redundant MFT (metadata), unless someone can explain me why NTFS's system is so much worse.
    NTFS has encryption since a long time ago. https://en.wikipedia.org/wiki/Encrypting_File_System
    NTFS has online fsck (in Win8 and Server 2012 and later). https://technet.microsoft.com/en-us/...(v=ws.11).aspx
    NTFS supports deduplication (although it is enabled only on server versions, you need to hack the Home/Pro versions to enable it afaik) http://gestaltit.com/syndicated/step...tfs-windows-8/

    NTFS added features well after that.
    You can add as many features as you like to NTFS it's still NTFS and it's still 24 years old. Same goes for EXT4, it's still basically EXT2 with a journal.

    Comment


    • #32
      Originally posted by arokh View Post
      Why can't you compare them again?
      By "you can't compare" people usually mean "it's an unfair comparison because you're comparing apples to bananas".

      And a full CoW filesystem with integrated checksumming AND raid is much different from a metadata-only CoW/checksumming without raid, as the former is far more complex.

      It's one of the reasons btrfs isn't yet ready while APFS is.

      Just because you don't see the point doesn't mean there isn't one. The keyword here is speed. Journaling has overhead.
      On metadata only? On SSDs? You think it matters? Did you see the benchmarks of ext4 on SSDs by phoronix?

      You can add as many features as you like to NTFS it's still NTFS and it's still 24 years old. Same goes for EXT4, it's still basically EXT2 with a journal.
      NTFS's main issue is that it is still crap, not that it is old or that it does not have features.
      For example EXT2 came out in 93 and after adding stuff it became EXT3 and then EXT4. It's still rocking in benchmarks.

      Comment


      • #33
        Originally posted by arokh View Post
        Why can't you compare them again? One has full CoW, the other hasn't. There, they are compared. Nobody said APFS matches BTRFS or ZFS feature by feature.
        Just because you don't see the point doesn't mean there isn't one. The keyword here is speed. Journaling has overhead.
        You can add as many features as you like to NTFS it's still NTFS and it's still 24 years old. Same goes for EXT4, it's still basically EXT2 with a journal.
        Age is relevant how? Last iteration is actually 16 years old. There have been 5 in total. On-disk format has remained unchanged since 2001, new releases of Windows's have just added functionalities (like symbolic links (debianxfce) - this bit is a news for you ), self-healing and partition shrinking).

        "When it works, don't fuck with it until it starts stopping your work". Principle, I understand, foreign to the tinkerer who just wants new and shiny thing which is supposedly "better" because it's "new". So, you'd end up with half-dozen semi-functional "better" things while best workhorses are still like 20+ years old (XFS, Ext iteration x)

        Comment


        • #34
          Originally posted by aht0 View Post
          Age is relevant how? Last iteration is actually 16 years old. There have been 5 in total. On-disk format has remained unchanged since 2001, new releases of Windows's have just added functionalities (like symbolic links (debianxfce) - this bit is a news for you ), self-healing and partition shrinking).

          "When it works, don't fuck with it until it starts stopping your work". Principle, I understand, foreign to the tinkerer who just wants new and shiny thing which is supposedly "better" because it's "new". So, you'd end up with half-dozen semi-functional "better" things while best workhorses are still like 20+ years old (XFS, Ext iteration x)
          Maybe you should ask Microsoft that, as they are replacing it with ReFS. It's already been answered in this thread and in every mailing list that covers new filesystems as well.

          Comment


          • #35
            Originally posted by arokh View Post

            Maybe you should ask Microsoft that, as they are replacing it with ReFS. It's already been answered in this thread and in every mailing list that covers new filesystems as well.
            "Is replacing". As "it is work-in-progress". It still lacks plenty of NTFS features. By the time it becomes production ready and is feature-complete, NTFS is probably slowly approaching the state of "it starts stopping your work". Because storage would have reached sizes where NTFS is just inefficient. Like 50+ TB of data. Some operations could take days there.

            ReFS is designed for very large data sets. Can you argue it's not something we could reasonably expect to come?

            Comment

            Working...
            X