Results 1 to 7 of 7

Thread: An Update On The Tux3 File-System

  1. #1
    Join Date
    Jan 2007
    Posts
    14,378

    Default An Update On The Tux3 File-System

    Phoronix: An Update On The Tux3 File-System

    Linux file-systems has been a hot topic as of late with the development of the Btrfs and Tux3 file-systems in particular. Tux3 is an open-source versioning file-system being developed by Daniel Phillips since earlier this year that builds upon the never-released Tux2...

    http://www.phoronix.com/vr.php?view=NjcyNQ

  2. #2
    Join Date
    Jun 2007
    Location
    The intarwebs
    Posts
    385

    Default

    I'm glad to see that the parts the user never sees are still getting solid amounts of attention. For the future!

  3. #3
    Join Date
    Sep 2007
    Location
    Paris, France
    Posts
    217

    Default

    Quote Originally Posted by ethana2 View Post
    I'm glad to see that the parts the user never sees are still getting solid amounts of attention. For the future!
    I'm glad too, but what is this filesystem ? I've never heard of that before. What is the advantage versus ext3 ? Why spending time developping a new filesystem and not concentrate on the ext4 filesystem ?

  4. #4
    Join Date
    May 2008
    Posts
    73

    Default

    Quote Originally Posted by Fixxer_Linux View Post
    Why spending time developping a new filesystem and not concentrate on the ext4 filesystem ?
    Well that's just how things work, you can't expect all people who are interested in developing filesystems to focus on one project.

    First because ext most likely has to maintain some level of backward compatibility, while you can try out whatever wild ideas you have if you start your own project.

    And second becuase after a while just putting more and more people on the same project just doesn't translate to better and faster progress, in fact quite the opposite might happen, the project might bog down because of internal problems.

    In the end it always works out, if Tux3 doesn't work it will just die without affecting existing projects. If it works really well, more and more people will start to use it and most likely existing projects will try to copy whatever makes it so succesful.

    That way either we will still be using ext5-6-7 in the future but with new ideas that were proven to be valuable by other projects like Tux3 or we might actually see a shift towards a completely new filesystem like tux3 instead of sticking with ext. Who knows. Either way the effort made these people is not lost (even if the project fails what they learned is valuable to others).

    (NB: I know there are other file systems besides ext, it's just that most distros seem to default to it, but I might be wrong here. Anyway it doesn't invalidate my point)

  5. #5
    Join Date
    Sep 2007
    Location
    Paris, France
    Posts
    217

    Default

    Quote Originally Posted by quintesse View Post
    Well that's just how things work, you can't expect all people who are interested in developing filesystems to focus on one project.

    First because ext most likely has to maintain some level of backward compatibility, while you can try out whatever wild ideas you have if you start your own project.

    And second becuase after a while just putting more and more people on the same project just doesn't translate to better and faster progress, in fact quite the opposite might happen, the project might bog down because of internal problems.

    In the end it always works out, if Tux3 doesn't work it will just die without affecting existing projects. If it works really well, more and more people will start to use it and most likely existing projects will try to copy whatever makes it so succesful.

    That way either we will still be using ext5-6-7 in the future but with new ideas that were proven to be valuable by other projects like Tux3 or we might actually see a shift towards a completely new filesystem like tux3 instead of sticking with ext. Who knows. Either way the effort made these people is not lost (even if the project fails what they learned is valuable to others).

    (NB: I know there are other file systems besides ext, it's just that most distros seem to default to it, but I might be wrong here. Anyway it doesn't invalidate my point)

    You're right. Smart answer. Anyway, that didn't give us what are the advantages of Tux3 against the rest of the world. Or did I missed something ?

  6. #6
    Join Date
    May 2008
    Posts
    73

    Default

    Quote Originally Posted by Fixxer_Linux View Post
    You're right. Smart answer. Anyway, that didn't give us what are the advantages of Tux3 against the rest of the world. Or did I missed something ?
    Well that's because I don't know anything about it's advantages

    Lot's of information here though: http://shapor.com/tux3/shapor-tux3/doc/design.html

    What I understand from rapidly scanning the text is that a journaling/versioning filesystem has a larger overhead (compared to a FS without those features) when making changes, the design of Tux3 tries to minimize that overhead.

  7. #7
    Join Date
    Sep 2007
    Location
    Paris, France
    Posts
    217

    Default

    Quote Originally Posted by quintesse View Post
    Well that's because I don't know anything about it's advantages
    Sorry not to have make myself more clear : I wasn't saying you should have given the advantages in your thread !
    I was saying that just in case someone else would have read the thread and would have known that in particular.
    Thank you very much for having searched the web for a little more information about that, that is very kind from you.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •