On Wed, Apr 30, 2008 at 12:51 PM, Sloan
G T Smith wrote:
Sloan wrote:
bleh, ext3 took over twice as long as either reiser or xfs.
Confirms my gut feel about the performance of the one (and probably only ever) ext3 partition which I set up as a bit of an experiment.
So, I continue with reiser 3, but looking down the road, reiser 4 seems a long shot at this point. I suppose it's possible that ext4 could turn out well. However Btrfs from oracle looks even more interesting to me -
Its my understanding that Reiserfs4's greatest improvements were to be in the area of file tagging so that you could apply virtually unlimited characteristics and attributes to files so they could be found more easily. This could be done automatically by software or manually. This allowed you to find files without the need of reading the entire directory, let alone the entire drive. Also improvements to the journaling of actual data instead of the Metadata-only journaling that Reiser 3 has. The bigger hard drives get, (and the more of them there are), the more this will become necessary. Vista promised to do this, but with tags were external (in another file), but ultimately never delivered. Right now, the world pretty much relies on "a place for everything, and everything in its place", with a secondary reliance on file type. This reliance on directory structures can be viewed as crutch needed by people; its how we think. But its not how disk drives actually work at the physical level. File type is closer to what we actually want, so that we don't try to listen to our pictures, or print out our executables. My understanding was that Reiserfs4 was going to simulate directory structures as simply one more attribute of a file, primarily for legacy sake. I'm not seeing any of those features in Btrfs. -- ----------JSA--------- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org