Journaled filesystem not suitable for large filesystems?

[cb911]

i was just reading up on Gentoo (going to install Getnoo 2004.0!! ) and came accross a bit of info about ext3 (the journaled version of the linux filesystem):

from the Gentoo handbook
ext3 is the journaled version of the ext2 filesystem, providing metadata journaling for fast recovery in addition to other enhanced journaling modes like full data and ordered data journaling. ext3 is a very good and reliable filesystem. It offers generally decent performance under most conditions. Because it does not extensively employ the use of "trees" in its internal design, it doesn't scale very well, meaning that it is not an ideal choice for very large filesystems, or situations where you will be handling very large files or large quantities of files in a single directory. But when used within its design parameters, ext3 is an excellent filesystem.

i was just wondering if this info about journaled filesystems applies to Panther and the way it handles journaling... i've never liked the journaling in Panther for some reason.

does anyone know if you had a G5 with a 250GB HD (single partiton with 10.3 installed) and journaling enabled, would that effect the performance?

 

[Felix_the_Mac]

Journalled file systems are intended for large filesystems.
They became necessary since doing an fsck (file system check) on a large modern disk can take hours whereas a jfs will take literally seconds.

The specific performance characteristics of any particular jfs implementation will differ, so the fact that EXT3 is said to be sub-optimal is not necessarily true of HFS+.

I don't know much more than that but I expect that if you google "journalled file systems HFS+ speed" etc etc. you could find some benchmarks.

 

[Flynnstone]

ext3 and ext2 are then same file system, ext2. What ext3 does is add journaling to ext2. A very simplistic look at journaling is this, a program via the filesystem is going to make a change to a file, so it writes to the journal what its about to do, then changes the file, then removes the entry in the journal.
So returning from my little tangent, it is said that ext2 doesn't scale well. So this implies ext3 doesn't scale well. So ext3 not scaling well is not specifically due to the journalling.
There have been some benchmark testing of ext3 and ext2 filesystems. It would seem logical that ext2 is faster than ext3 due to the extra journalling action of ext3. But ... there are cases where ext3 is faster than ext2.

But I think it is safe to say that ext3 is inheritantly "safer" than ext2, ie your data better protected.
Now HFS+ vs ext2 vs ext3 ... I don't know that answer, but in my opinion, journalling is better.

 

[superbovine]

gee i learned a lot

 

[cb911]

okay, i understand it now. didn't think about the differences between ext2/ext3 and HFS+.

i remember reading all about the way Panther uses journaling just to prove to myself that it wasn't going to stuff everything up. but i always forget about the details. maybe i do need a refresher.