Re: New filesystem for Linux

[Oleg Verych <olecom@flower.upol.cz>]

On 2006-11-03, Mikulas Patocka wrote:
[]
>> Well, at least for XFS everybody tell that it should be used with UPS only if 
>> you really care about your data. I think it has something to do with heavy 
>> in-RAM caching this filesystem does.
>
> System is allowed to cache anything unless sync/fsync is called. Someone 
> told that XFS has some bugs that if crashed incorrectly, it can lose 
> already synced data ... don't know. Plus it has that infamous feature (not 
> a bug) that it commits size-increase but not data and you see zero-filed 
> files.

AFAIK, XFS by design must provide _file system_ consistency, not data.

[]
>> PS. Do you have any benchmarks of your filesystem? Did you do any longer 
>> automated tests to prove it is not going to loose data to easily?
>
> I have, I may find them and post them. (but the university wants me to 
> post them to some conference, so I should keep them secret :-/)

Interesting to know, how "another one" FS is quick and stable.

On my new hardware, with dual core CPU 3.4G, 1G of RAM, 1/2TB disk space
(office pro, yes *office* (pro)), `dd' can suck 50M/s, (running and)
extracting 2.6.19-rc2 on fresh 20Gb partition (xfs,jfs) yields nearly 4M/s.

As for ordinary user it seems very slowly.

Ext2 (not ext3) is as fast as shmfs, until RAM will be full.
And after 13 cycles XFS has 11-12 directories with good linux source,
ext2 6-7 (IIRC). On start of 14th cycle i pushed reset button, btw.

Mounting & repairing XFS took less than minute; e2fsck spent more time
on printing, rather than reparing, i think (:.

Want to create another one? OK, but why not to improve existing? >/dev/null
____