Announcement

Collapse
No announcement yet.

EXT4 Lands A Nice Performance Improvement For Appending To Delalloc Files

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

  • #21
    Originally posted by mrg666 View Post
    There are multiple other options but I see no reason to try another file system as ext4, and 2, 3 before that, have all been very reliable and fast for my workstations. Compiling the kernel, for example, is never limited by the file system as CPU load is always close to 100%. I don't think Abaqus or GIMP will be any faster with another file system either. It is good to see improvements are still being done on such a mature software.
    If FS-level compression/dedupe/redundancy doesn't get your motor running then there is no need. I wish BTRFS could do extension-based compression because then it would be much less painful to do high compression on things that can compress. The current state of play is that unless you force compression it'll turn off compression once enough incompressible data has been found, not so useful for tar/iso/archives that can mix arbitrarily.

    Comment


    • #22
      Originally posted by geerge View Post
      If FS-level compression/dedupe/redundancy
      How do you use de-duplication? The only way I'm aware of it being used is in snapshots.

      As I've mentioned before, I view checksums as the main value that I derive from BTRFS, on my personal machines.

      Comment


      • #23
        Originally posted by coder View Post
        How do you use de-duplication? The only way I'm aware of it being used is in snapshots.

        As I've mentioned before, I view checksums as the main value that I derive from BTRFS, on my personal machines.

        Comment


        • #24
          Originally posted by skeevy420 View Post

          When I first came here, this was all swamp. Everyone said I was daft to build an extent based file system on a swamp, but I built in all the same, just to show them. It sank into the swamp. So I built a second one. That sank into the swamp. So I built a third. That burned down, fell over, then sank into the swamp. But the fourth one stayed up. And that's what you're going to get, Lad, the strongest extent based file system in all of England.
          ext2 and ext3 don't support extents at all. The "ext" stands for extended and only until the fourth extended file system the extent feature were added

          Comment


          • #25
            Originally posted by david-nk View Post
            So basically a log file optimization?
            Trivially, yes. But it's not as trivial as it sounds, since there's lots of applications using log-structured file formats to get log-structured filesystem performance benefits from non-log-structured filesystems, and/or their use-case involves constant incremental data being stored without too much worry around running out of disk space. An early example of this still in use today would be the mbox file format. SQLite is an example of one you're using a lot today, as it's used by Firefox, Google Chrome, and Safari to maintain session state for tabs, as well as cookies and the Web Storage API, and SQLite is getting (IMHO) over-adopted as a bloated per-app registry in place of more robust and user-readable config files. Databases in general end up acting like logs even where they aren't designed from the ground up to get the best benefits from being logs.
            Last edited by linuxgeex; 06 September 2023, 08:06 PM.

            Comment

            Working...
            X