Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Filesystems comparison for present time (r4,r3,jfs,xfs,ext3)
View unanswered posts
View posts from last 24 hours

Goto page 1, 2, 3, 4, 5  Next  
Reply to topic    Gentoo Forums Forum Index Off the Wall
View previous topic :: View next topic  

Please READ the tables,next for which fs You want to switch ? :)
ext3
26%
 26%  [ 68 ]
jfs
10%
 10%  [ 26 ]
reiser3
17%
 17%  [ 44 ]
reiser4
35%
 35%  [ 91 ]
xfs
9%
 9%  [ 25 ]
Total Votes : 254

Author Message
fallow
Bodhisattva
Bodhisattva


Joined: 08 Jan 2004
Posts: 2206
Location: Poland

PostPosted: Tue Apr 12, 2005 10:51 am    Post subject: Filesystems comparison for present time (r4,r3,jfs,xfs,ext3) Reply with quote

Hey.

Yeah . Propably many of You like reiser4 and etc. This thread is not to discuss who are liking it and the reason of liking it or not liking it :) Im a user of filesystem . I did a simple test to compare a simple things which I doing day by day on my box.

Why ?
I was used reiser4 for a long time. And I was not happy because of some things. I Switched to reiser3. Im not 100% happy because of high cpu usage with it. (dont ask me for reiser4 ;) . In the past I used ext2 and ext3 with the redhat for a long time, also jfs and xfs on Gentoo. JFS code has some improvements and I wanted to see what is the current situation at present time.

so . this is a results .

my box : AMD 1.0 Athlon , 256 MB Ram
HDD used for tests IBM
hdd: IC35L040AVVA07-0, ATA DISK drive
hdd: max request size: 128KiB
hdd: 80418240 sectors (41174 MB) w/1863KiB Cache, CHS=65535/16/63, UDMA(100)
all tests with the same kernel , IngoSched ,CFQ-TS IO Scheduler , and latest versions of all fs code.

1. FIRST - bonnie++ test (remember that us is smaller than ms , so better in latency )
Code:

-------------------------------------------------------------------------------
bonnie++ -u root /mnt/t

Version 1.93c       ------Sequential Output------ --Sequential Input- --Random-
Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
reiser4        496M    83  98 33536  23 17221  20   120  99 25225  32 185.8  10
reiser3        496M   167  99 33760  28 16974  13   594  99 38790  17 191.4   7
jfs            496M   399  97 34165  19 17232  10   636  98 38967  14 192.5   5
xfs            496M   228  98 35612  20 16558  12   416  98 39085  16 116.1   3
ext3           496M   139  99 33906  33 16618  13   551  98 38555  15 187.8   6

reiser4-lat           157ms    2749ms     874ms   76953us   10684us    3401ms
reiser3-lat         64610us    1231ms     226ms   23227us   22063us     963ms
jfs-lat             24487us     655ms     160ms   42963us   99906us     708ms
xfs-lat             39552us    1954ms     159ms   47486us   35691us     554ms
ext3-lat            67131us    1215ms     249ms   48126us   99742us     798ms
-------------------------------------------------------------------------------
Version 1.93c       ------Sequential Create------ --------Random Create--------
Enterprise          -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
reiser4          16 17829  90 +++++ +++  8190  98  8174  95 +++++ +++  8307  98
reiser3          16 11570  89 +++++ +++ 10999  98 11402  91 +++++ +++  9970  98
jfs              16  3569  16 +++++ +++  2177  11  1349  17 +++++ +++  1023   9
xfs              16  2461  35 +++++ +++  2203  27  2429  35 +++++ +++   884  13
ext3             16   816  98 +++++ +++ 24692  77   780  97 +++++ +++  2256  96

reiser4-lat          2820us    8955us    9138us    2976us     127us    3365us
reiser3-lat          8458us    3623us    4707us    3648us     134us    2846us
jfs-lat             61168us    1374us     182ms     248ms     134us    1491ms
xfs-lat               118ms     291us     119ms     108ms     215us     135ms
ext3-lat            37327us     205us    2779us   36279us    1161us   65357us
-------------------------------------------------------------------------------


Now a copying kernel_sources and deleting it test
Code:

time cp  -r /usr/src/linux-2.6.12-rc2 /mnt/t

reiser4 real    0m49.015s
reiser3 real    1m1.828s
jfs     real    0m58.990s
xfs     real    1m1.678s
ext3    real    0m48.338s

time rm -r /mnt/t/linux-2.6.12-rc2
reiser4 real    0m7.663s
reiser3 real    0m2.308s
jfs     real    0m13.117s
xfs     real    0m16.872s
ext3    real    0m1.896s


3. and copying and deleting some livecd also
Code:

time cp -r /hdd-livecd/ /mnt/t

reiser4 real    1m3.217s
reiser3 real    1m1.387s
jfs     real    1m23.267s
xfs     real    1m26.931s
ext3    real    1m1.526s

time rm -r /mnt/t/hdd-livecd/
reiser4 real    0m11.666s
reiser3 real    0m4.669s
jfs     real    0m38.186s
xfs     real    0m32.916s
ext3    real    0m7.207s


some quick resume :

1.Want to discover the America once again ?
EXT3 is a most universal filesystem for overall using
the speed of copying kernel and livecd files is the same as with reiser4
every other fs is slower.
Depending of high cpu usage by reiser*** filesystems ( Yes , I want to have better interactivity from the system ) I decided to move to the "roots" .

I see 2 alternatives now : EXT3 and JFS .

EXT3 is more universal and average . JFS _IN OVERALL_ is little slower than EXT3 but has incredible _LOW_ cpu usage in some cases.

IM undecided then .... and dont know what to choose EXT3 or JFS ..

please report Your opinion about ext3 and jfs comparison

cheers
fallow
_________________
"Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard ;)


Last edited by fallow on Fri Apr 15, 2005 4:13 pm; edited 2 times in total
Back to top
View user's profile Send private message
Enlight
Advocate
Advocate


Joined: 28 Oct 2004
Posts: 3514
Location: Alsace (France)

PostPosted: Tue Apr 12, 2005 11:33 am    Post subject: Reply with quote

Well currently I'm an xfs user, and I was thinking that it was time using it a better way, it'seems that there are somme great features available like putting journalisation on another hd than the one you're writting on wich should massively improve latency (I was thinking about getting an old hd for stuffs like this) and I've heard that there were ways to makes it perform very fast on small files.

But now your post make me want to give a try at JFS, but I was wondering if multiples files sytem support won't perform slower (ie using reiser3.6 for /etc, xfs for /home...) for 1) making kernel bigger and 2) increasing cpu load?
Back to top
View user's profile Send private message
Cintra
Advocate
Advocate


Joined: 03 Apr 2004
Posts: 2111
Location: Norway

PostPosted: Tue Apr 12, 2005 11:42 am    Post subject: Reply with quote

Hei
I currently use reiser3.6 but I'm thinking of putting up a test 2005 partition using jfs
the following post helped make up my mind..
Quote:
cybermonkey
Tux's lil' helper
Joined: 11 Aug 2003
Posts: 84
PostPosted: Sun Apr 10, 2005 5:22 pm

I've been using JFS on my everyday comp now for about 4 months and have had no problems whatsoever Very Happy

The reason i use it is because everynow and then the power company around here like to switch this area off at will and ..

JFS recovers in a snap with all data intact


Oh, and thanks for the comparison effort

mvh
_________________
"I am not bound to please thee with my answers" W.S.
Back to top
View user's profile Send private message
fallow
Bodhisattva
Bodhisattva


Joined: 08 Jan 2004
Posts: 2206
Location: Poland

PostPosted: Tue Apr 12, 2005 12:00 pm    Post subject: Reply with quote

Hmm. Yeah . journal data on other hd can be a very good idea. but I dont have possibilities to do it :(

my target is a home desktop

In the past I had many of different fs for /usr , portage etc. but now I want to stick with the one.

I think that ext3 can be a best choice.

cheers.
_________________
"Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard ;)
Back to top
View user's profile Send private message
IvanYosifov
l33t
l33t


Joined: 15 Oct 2004
Posts: 778
Location: Bulgaria

PostPosted: Tue Apr 12, 2005 12:19 pm    Post subject: Reply with quote

fallow, it may be my browser fonts, but the text in the table is missaligned :( Can you please put the data in a plain text file an upload it somewhere ?
Back to top
View user's profile Send private message
Enlight
Advocate
Advocate


Joined: 28 Oct 2004
Posts: 3514
Location: Alsace (France)

PostPosted: Tue Apr 12, 2005 12:29 pm    Post subject: Reply with quote

Yes I think about sticking with only one too, in order to have an as small as possible kernel, that's why I was wondering If stuffs like initrd could be done with xfs as an example, putting journalisation elsewhere.

BTW I was thinking for desktop too but as I 've got a sata disk I was thinking about getting an old and cheap little ide disk to put journalisation on it, or why not an usb key... or something insane like this :D

ps : I'm currently trying vivids which seemed to be the closest sources from what I was looking for, and would soon report back my impressions on 'em


Last edited by Enlight on Tue Apr 12, 2005 12:42 pm; edited 1 time in total
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Tue Apr 12, 2005 12:35 pm    Post subject: ext3 Reply with quote

For a while, I had decided to switch to ext3 from reiser3, and found a major issue, at times, when moving or copying a bunch of < 25kb files, the file system would lock up, and cause me to reboot, this has actually happened to me on more than one occassion, with gentoo, and with slackware. With reiser3, i never have this issue. I am curious about JFS though.
Back to top
View user's profile Send private message
Sith_Happens
Veteran
Veteran


Joined: 15 Dec 2004
Posts: 1807
Location: The University of Maryland at College Park

PostPosted: Tue Apr 12, 2005 1:52 pm    Post subject: Reply with quote

Here is a pretty good benchmark I found a while ago for xfs, jfs, xext3, ext2, and reiserfs 3.6. Before you say the benchmark is too old, look at the hardware they are using. I think your benchmarks are good fallow, just check these out too. Anyway I use Reiserfs 3.6, I think it has the best mix of stability and speed for my application. However, if you look at the benchmark, you'll see that each filesystem has it's advantages and disadvantages, it's not really like there is a "best filesystem" for all purposes. However, it does seem that this benchmark shows there is worst filesystem for all applications ext3. :wink: We'll, not for all aplications, it is stable as hell, but it's also as slow as it is stable. :lol:
_________________
"That question was less stupid; though you asked it in a profoundly stupid way."
I'm the brains behind Jackass! | Tutorials: Shorewall
Back to top
View user's profile Send private message
Enlight
Advocate
Advocate


Joined: 28 Oct 2004
Posts: 3514
Location: Alsace (France)

PostPosted: Tue Apr 12, 2005 3:14 pm    Post subject: Reply with quote

Quote:
LIBC VERSION: 2.3.2
KERNEL: linux-2.4.26
COMPILER USED: gcc-3.3.3
EXT2: e2fsprogs-1.35/sbin/mkfs.ext2
EXT3: e2fsprogs-1.35/sbin/mkfs.ext3
JFS: jfsutils-1.1.5/sbin/mkfs.jfs
REISERFS: reiserfsprogs-3.6.14/sbin/mkreiserfs
XFS: xfsprogs-2.5.6/sbin/mkfs.xfs


Actually there have been recent improvement into JFS, and the test takes ext3 and xfs stock, as I said before, there are probably major improvements that can be done both xfs, and some could be done with ext3 too (see codergeek's post for ext3).


BTW if someone could explain me what real time allocator does and how it performs...

Quote:
Optional Realtime Allocator

XFS supports the notion of a "realtime subvolume" - a separate area of disk space where only file data is stored. Space on this subvolume is managed using the realtime allocator (as opposed to the default, B+ tree space allocator). The realtime subvolume is designed to provide very deterministic data rates suitable for media streaming applications.
Back to top
View user's profile Send private message
Sith_Happens
Veteran
Veteran


Joined: 15 Dec 2004
Posts: 1807
Location: The University of Maryland at College Park

PostPosted: Tue Apr 12, 2005 4:52 pm    Post subject: Reply with quote

Code:
Optional Realtime Allocator

XFS supports the notion of a "realtime subvolume" - a separate area of disk space where only file data is stored. Space on this subvolume is managed using the realtime allocator (as opposed to the default, B+ tree space allocator). The realtime subvolume is designed to provide very deterministic data rates suitable for media streaming applications.
Yeah, I saw this too. Right now I'm using xfs for a filehosting application, and I would be interested in what I could use this aspect of xfs for in that area.
_________________
"That question was less stupid; though you asked it in a profoundly stupid way."
I'm the brains behind Jackass! | Tutorials: Shorewall
Back to top
View user's profile Send private message
ewan.paton
Veteran
Veteran


Joined: 29 Jul 2003
Posts: 1219
Location: glasgow, scotland

PostPosted: Tue Apr 12, 2005 5:29 pm    Post subject: Reply with quote

i uses xfs for /home reiserfs for / and may get round to switching /usr/portage to reiser4 eventualy i cant quite place why i like xfs for big files but i do, i think its about matching whats best for the job
Back to top
View user's profile Send private message
Illissius
Guru
Guru


Joined: 31 Jul 2004
Posts: 395
Location: Hungary

PostPosted: Tue Apr 12, 2005 5:36 pm    Post subject: Reply with quote

(You did reboot or otherwise clear the filecache between tests, right?)

Thanks a lot for the numbers. For JFS's and XFS's low performance and CPU utilization during the file create/remove tests, the explanation is right below: incredibly high latency. This is not a good thing, it means system resources can't being fully utilized. EXT3 also has very high latency (though not as high as X/JFS), but that apparently doesn't prevent it from consuming 100% CPU while performing piss poorly. These numbers also explain why reiser* is so fast at emerge syncing, while not necessarily at other things: they're very fast at file creation/deletion, which emerge sync apparently involves lots of. With the exception of reiser4 rather underperforming, the data read/write tests don't seem to have much variance, which makes sense, as I would assume there's less filesystem involvement anyways (filesystems manage the stuff between the files, not so much the stuff in them).
I'm also surprised about the kernel source and livecd copy/remove test results... the traditionally expected results would have been reiser* dominating the copy/remove tests, and XFS likewise for the livecd copying, which plainly did not happen.

Anyways, the conclusions I would draw from this data (the results would have to be independently verified before I'd put any trust in it):
- reiser4 doesn't (yet) live up to the hype. at best it manages to keep pace with the others, at worst it's significantly off it. coupled with the potential stability issues, it doesn't look too attractive.
- XFS and JFS suck at both file creation/deletion and copying large files, while not making up for it in the other tests. at least they are likely to actually *work*, but nothing to get excited about. (such high latency and poor performance for XFS in copying a large file make me think that there may have been issues with the test configuration in fact, hence the disclaimer above...)
- EXT3 is fucking abysmal at creating/removing files, but performs very solidly at everything else. Assuming you don't mind your emerge syncs going slower (is there anything else that involves heavy file creation/removal besides that?), and given its solid track record, it's a fine enough choice.
- reiser3 leaves everything else in its dust for creating and removing files (only reiser4 gets close), and keeps up quite well in all the other tests. and has a likewise solid track record. it would, hence, be my pick.
(Again, let me reiterate that these conclusions are based on, and only on, the data presented above. More data (such as XFS actually not sucking at the thing it was meant to be good at) may well end up changing them.)

And thanks again to fallow for making the data available.
_________________
Work is punishment for failing to procrastinate effectively.
last.fm


Last edited by Illissius on Tue Apr 12, 2005 5:45 pm; edited 1 time in total
Back to top
View user's profile Send private message
Lechium
Apprentice
Apprentice


Joined: 04 Apr 2005
Posts: 244

PostPosted: Tue Apr 12, 2005 5:38 pm    Post subject: Reply with quote

This is your choice, really... I've used all of these aside JFS, and honestly couldn't tell the difference between ext2 and reiserfs bu simply using these. Tests and such show the descrepency, but as end user experience goes the differences are miniscule.
On the side note -- pick whatever you like as long as it has journaling. Journaling will slow things down (by a hardly noticeble amount), but wil avoid possible half-writen files etc, so IHMO its worth it.

P.S. reiserfs4 can be the best choice for a simple reason -- a lot of people use it, so if you get into trouble it will be easier finding help, than if you run jfs/xfs.
Back to top
View user's profile Send private message
codergeek42
Bodhisattva
Bodhisattva


Joined: 05 Apr 2004
Posts: 5142
Location: Anaheim, CA (USA)

PostPosted: Tue Apr 12, 2005 5:44 pm    Post subject: Reply with quote

Illissius wrote:
- EXT3 is fucking abysmal at creating/removing files, but performs very solidly at everything else. Assuming you don't mind your emerge syncs going slower (is there anything else that involves heavy file creation/removal besides that?), and given its solid track record, it's a fine enough choice.
Actually, ext3 has very good I/O when reading/writing lots of stuff simultaneously (including many creates/writes/deletions in succession) if you enable full data journalling (not simply use the default metadata journalling and block grouping). But I do understand that you're making this conclusion from the data, so this would not be a valid argument in this case to use ext3.


Anyways, from what you've given us fallow, and from what I've read I'm actually very interested in how JFS can perform on a desktop/workstation box as compared to ext3 or ReiserFS v3.
_________________
~~ Peter: Brony, GNU/Linux geek, caffeine addict, and Free Software advocate.
Who am I? :: EFF & FSF
Back to top
View user's profile Send private message
i92guboj
Moderator
Moderator


Joined: 30 Nov 2004
Posts: 9899
Location: Córdoba (Spain)

PostPosted: Tue Apr 12, 2005 5:51 pm    Post subject: Reply with quote

Lechium wrote:
This is your choice, really... I've used all of these aside JFS, and honestly couldn't tell the difference between ext2 and reiserfs bu simply using these. Tests and such show the descrepency, but as end user experience goes the differences are miniscule.
On the side note -- pick whatever you like as long as it has journaling. Journaling will slow things down (by a hardly noticeble amount), but wil avoid possible half-writen files etc, so IHMO its worth it.

P.S. reiserfs4 can be the best choice for a simple reason -- a lot of people use it, so if you get into trouble it will be easier finding help, than if you run jfs/xfs.


1.- I doubt it is the most used fs in linux, since it is new and not in the kernel. I don't have official info or something like that, but I think the most used linux fs could be one of ext3 or reiserfs 3.6.

2.- If there are lots of info in the forums about it is because if fails and then people ask for help. Ext3 never fails, so no need to ask.

3.- Paraphrasing: 'windows xp can be the best choice for a simple reason -- a lot (much more than reiser4) of people use it, so if you get into trouble it will be aesier finding help, than if you run linux." :twisted:

Dont bother, just jokinng :wink:
_________________
Gentoo Handbook | My website
Back to top
View user's profile Send private message
Lechium
Apprentice
Apprentice


Joined: 04 Apr 2005
Posts: 244

PostPosted: Tue Apr 12, 2005 5:57 pm    Post subject: Reply with quote

6thpink wrote:
Lechium wrote:
This is your choice, really... I've used all of these aside JFS, and honestly couldn't tell the difference between ext2 and reiserfs bu simply using these. Tests and such show the descrepency, but as end user experience goes the differences are miniscule.
On the side note -- pick whatever you like as long as it has journaling. Journaling will slow things down (by a hardly noticeble amount), but wil avoid possible half-writen files etc, so IHMO its worth it.

P.S. reiserfs4 can be the best choice for a simple reason -- a lot of people use it, so if you get into trouble it will be easier finding help, than if you run jfs/xfs.


1.- I doubt it is the most used fs in linux, since it is new and not in the kernel. I don't have official info or something like that, but I think the most used linux fs could be one of ext3 or reiserfs 3.6.

2.- If there are lots of info in the forums about it is because if fails and then people ask for help. Ext3 never fails, so no need to ask.

3.- Paraphrasing: 'windows xp can be the best choice for a simple reason -- a lot (much more than reiser4) of people use it, so if you get into trouble it will be aesier finding help, than if you run linux." :twisted:

Dont bother, just jokinng :wink:


ext2/3 are the most used linux file systems, however as I said above -- file journaling is a very very good thing to have.

P.S. I have absolutley no animosity towards NT4.0, as Windows goes =) (oh damn I'm about to get flamed...)
Back to top
View user's profile Send private message
fallow
Bodhisattva
Bodhisattva


Joined: 08 Jan 2004
Posts: 2206
Location: Poland

PostPosted: Tue Apr 12, 2005 6:02 pm    Post subject: Reply with quote

Im little bus at the moment , and not reading all but

is the one thing - > speed of filesystem isnt all . In most cases faster fs occupied this by lost of interactivity . Is important to rethinnk what do U want - faster FS at all costs , or good overall interactivity . Im trying to find a good balance beetwen everything now .

and I think that is must to have better testing than this . this is _very_ very simple test.
JUST to start a discussion . Is many of other better methods . (OH sorry bonnie++ results was good )

[b] ANOTHER TEST [b]
One of friends from Polish forums did testing based on the good this ( http://linuxgazette.net/102/piszcz.html ) test.

WHAT IS WHAT wrote:

001] Create 10,000 files with touch in a directory.
002] Run 'find' on that directory.
003] Remove the directory.
004] Create 10,000 directories with mkdir in a directory.
005] Run 'find' on that directory.
006] Remove the directory containing the 10,000 directories.
007] Copy kernel tarball from other disk to test disk.
008] Copy kernel tarball from test disk to other disk.
009] Untar kernel tarball on the same disk.
010] Tar kernel tarball on the same disk.
011] Remove kernel source tree.
012] Copy kernel tarball 10 times.
013] Create 1GB file from /dev/zero.
014] Copy the 1GB file on the same disk.
015] Split a 10MB file into 1000 byte pieces.
016] Split a 10MB file into 1024 byte pieces.
017] Split a 10MB file into 2048 byte pieces.
018] Split a 10MB file into 4096 byte pieces.
019] Split a 10MB file into 8192 byte pieces.
020] Copy kernel source tree on the same disk.
021] Cat a 1GB file to /dev/null.
RESULTS
Code:

Filesystem: ext2
 - creating filesystem with options '-L test':   4.375 seconds
 - mounting filesystem:   0.034 seconds
 - filesystem info:
     /dev/evms/test        10321204        20   9796900   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type ext2 (rw)
 - testing filesystem:
   + test T01:  14.950 seconds,       668.90 files/second
   + test T02:   0.013 seconds,    794721.94 files/second
   + test T03:   0.078 seconds,    127751.65 files/second
   + test T04:  26.667 seconds,       375.00 dirs/second
   + test T05:   0.171 seconds,     58426.26 dirs/second
   + test T06:   0.403 seconds,     24814.02 dirs/second
   + test T07:   1.965 seconds,  18589993.41 bytes/second
   + test T08:   0.983 seconds,  37160948.18 bytes/second
   + test T09:  34.229 seconds
   + test T10:  26.639 seconds
   + test T11:   0.270 seconds
   + test T12:  19.772 seconds
   + test T13:  23.562 seconds,  45570793.69 bytes/second
   + test T14:  50.427 seconds,  21292837.24 bytes/second
   + test T15:   5.769 seconds
   + test T16:   4.992 seconds
   + test T17:   1.482 seconds
   + test T18:   0.494 seconds
   + test T19:   0.229 seconds
   + test T20:   8.233 seconds
   + test T21:  22.315 seconds
 - unmounting filesystem:   0.182 seconds

Filesystem: ext3
 - creating filesystem with options '-L test -j':   5.526 seconds
 - mounting filesystem:   0.086 seconds
 - filesystem info:
     /dev/evms/test        10321204     32828   9764092   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type ext3 (rw)
 - testing filesystem:
   + test T01:  21.473 seconds,       465.70 files/second
   + test T02:   0.069 seconds,    145368.04 files/second
   + test T03:   0.231 seconds,     43289.83 files/second
   + test T04:  24.928 seconds,       401.16 dirs/second
   + test T05:   0.191 seconds,     52455.16 dirs/second
   + test T06:   1.188 seconds,      8417.65 dirs/second
   + test T07:   2.018 seconds,  18100318.74 bytes/second
   + test T08:   0.928 seconds,  39376133.38 bytes/second
   + test T09:  36.182 seconds
   + test T10:  25.859 seconds
   + test T11:   0.544 seconds
   + test T12:  20.311 seconds
   + test T13:  25.038 seconds,  42884207.81 bytes/second
   + test T14:  53.003 seconds,  20258278.62 bytes/second
   + test T15:   7.939 seconds
   + test T16:   9.878 seconds
   + test T17:   2.096 seconds
   + test T18:   0.624 seconds
   + test T19:   0.278 seconds
   + test T20:  19.462 seconds
   + test T21:  22.321 seconds
 - unmounting filesystem:   0.196 seconds

Filesystem: jfs
 - creating filesystem with options '-L test -q':   0.874 seconds
 - mounting filesystem:   0.183 seconds
 - filesystem info:
     /dev/evms/test        10452464      1408  10451056   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type jfs (rw)
 - testing filesystem:
   + test T01:  12.114 seconds,       825.51 files/second
   + test T02:   0.054 seconds,    183516.40 files/second
   + test T03:   0.905 seconds,     11045.53 files/second
   + test T04:  12.001 seconds,       833.27 dirs/second
   + test T05:   0.291 seconds,     34322.29 dirs/second
   + test T06:   1.652 seconds,      6051.72 dirs/second
   + test T07:   1.999 seconds,  18276310.88 bytes/second
   + test T08:   0.922 seconds,  39616427.42 bytes/second
   + test T09:  36.251 seconds
   + test T10:  26.824 seconds
   + test T11:   4.377 seconds
   + test T12:  18.882 seconds
   + test T13:  26.052 seconds,  41215360.39 bytes/second
   + test T14:  88.937 seconds,  12073091.94 bytes/second
   + test T15:   3.477 seconds
   + test T16:   3.367 seconds
   + test T17:   1.360 seconds
   + test T18:   0.548 seconds
   + test T19:   0.347 seconds
   + test T20:  23.082 seconds
   + test T21:  22.170 seconds
 - unmounting filesystem:   0.246 seconds

Filesystem: xfs
 - creating filesystem with options '-L test -f':   0.683 seconds
 - mounting filesystem:   0.217 seconds
 - filesystem info:
     /dev/evms/test        10475456       272  10475184   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type xfs (rw)
 - testing filesystem:
   + test T01:  12.756 seconds,       783.93 files/second
   + test T02:   0.050 seconds,    198601.46 files/second
   + test T03:   1.750 seconds,      5714.73 files/second
   + test T04:  12.863 seconds,       777.41 dirs/second
   + test T05:   0.165 seconds,     60710.55 dirs/second
   + test T06:   1.447 seconds,      6909.05 dirs/second
   + test T07:   1.974 seconds,  18502968.61 bytes/second
   + test T08:   0.785 seconds,  46523469.80 bytes/second
   + test T09:  37.182 seconds
   + test T10:  25.806 seconds
   + test T11:   1.771 seconds
   + test T12:  19.094 seconds
   + test T13:  22.615 seconds,  47479456.22 bytes/second
   + test T14:  48.297 seconds,  22232217.43 bytes/second
   + test T15:   4.262 seconds
   + test T16:   3.750 seconds
   + test T17:   1.942 seconds
   + test T18:   0.995 seconds
   + test T19:   0.510 seconds
   + test T20:  19.657 seconds
   + test T21:  22.227 seconds
 - unmounting filesystem:   0.196 seconds

Filesystem: reiserfs
 - creating filesystem with options '-l test -q':   0.940 seconds
 - mounting filesystem:   0.847 seconds
 - filesystem info:
     /dev/evms/test        10485372     32840  10452532   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type reiserfs (rw)
 - testing filesystem:
   + test T01:  11.278 seconds,       886.66 files/second
   + test T02:   0.061 seconds,    163875.85 files/second
   + test T03:   0.407 seconds,     24583.01 files/second
   + test T04:  11.098 seconds,       901.05 dirs/second
   + test T05:   0.365 seconds,     27372.05 dirs/second
   + test T06:  11.230 seconds,       890.44 dirs/second
   + test T07:   1.982 seconds,  18433120.28 bytes/second
   + test T08:   0.969 seconds,  37721601.34 bytes/second
   + test T09:  36.846 seconds
   + test T10:  25.942 seconds
   + test T11:   1.170 seconds
   + test T12:  20.710 seconds
   + test T13:  25.038 seconds,  42884903.25 bytes/second
   + test T14:  59.869 seconds,  17934932.70 bytes/second
   + test T15:   1.390 seconds
   + test T16:   1.240 seconds
   + test T17:   0.752 seconds
   + test T18:   0.357 seconds
   + test T19:   0.257 seconds
   + test T20:  27.548 seconds
   + test T21:  24.886 seconds
 - unmounting filesystem:   0.230 seconds

Filesystem: reiser4
 - creating filesystem with options '-L test -y':   0.376 seconds
 - mounting filesystem:   0.136 seconds
 - filesystem info:
     /dev/evms/test         9963520       460   9963060   1% /mnt/testfs
     /dev/evms/test on /mnt/testfs type reiser4 (rw)
 - testing filesystem:
   + test T01:  11.642 seconds,       858.97 files/second
   + test T02:   0.101 seconds,     99040.22 files/second
   + test T03:   0.741 seconds,     13501.23 files/second
   + test T04:  11.870 seconds,       842.48 dirs/second
   + test T05:   0.260 seconds,     38405.83 dirs/second
   + test T06:  30.898 seconds,       323.65 dirs/second
   + test T07:   1.968 seconds,  18565535.15 bytes/second
   + test T08:   0.944 seconds,  38694528.41 bytes/second
   + test T09:  38.743 seconds
   + test T10:  26.569 seconds
   + test T11:   2.351 seconds
   + test T12:  17.703 seconds
   + test T13:  23.650 seconds,  45401134.44 bytes/second
   + test T14:  48.254 seconds,  22251975.55 bytes/second
   + test T15:   1.140 seconds
   + test T16:   0.836 seconds
   + test T17:   0.546 seconds
   + test T18:   0.414 seconds
   + test T19:   0.310 seconds
   + test T20:  14.530 seconds
   + test T21:  22.834 seconds
 - unmounting filesystem:   0.996 seconds


http://wa.fema.pl/~gkowal/fsbench/

now we have another thing to interprete : I THINK THIS test is better much hehe :)
_________________
"Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard ;)
Back to top
View user's profile Send private message
codergeek42
Bodhisattva
Bodhisattva


Joined: 05 Apr 2004
Posts: 5142
Location: Anaheim, CA (USA)

PostPosted: Tue Apr 12, 2005 6:20 pm    Post subject: Reply with quote

Lechium wrote:
ext2/3 are the most used linux file systems, however as I said above -- file journaling is a very very good thing to have.
And that's the exact purpose of ext3. It takes the tried and proven ext2 filesystem and adds a JBD (Journalling Block Device) layer which theoretically allows journalling on any block device I/O. To my understandning though this JBD is still only used in ext3, while the other journalling FSes use their own journalling code.
Quote:
P.S. I have absolutley no animosity towards NT4.0, as Windows goes =) (oh damn I'm about to get flamed...)
Yes you are :twisted:
_________________
~~ Peter: Brony, GNU/Linux geek, caffeine addict, and Free Software advocate.
Who am I? :: EFF & FSF
Back to top
View user's profile Send private message
Illissius
Guru
Guru


Joined: 31 Jul 2004
Posts: 395
Location: Hungary

PostPosted: Tue Apr 12, 2005 7:17 pm    Post subject: Reply with quote

fallow wrote:
[b] ANOTHER TEST [b]
One of friends from Polish forums did testing based on the good this ( http://linuxgazette.net/102/piszcz.html ) test.

...snip...

now we have another thing to interprete : I THINK THIS test is better much hehe :)

8O
anyone with the requisite skillz care to make some nice, digestable graphs out of that?
_________________
Work is punishment for failing to procrastinate effectively.
last.fm
Back to top
View user's profile Send private message
i92guboj
Moderator
Moderator


Joined: 30 Nov 2004
Posts: 9899
Location: Córdoba (Spain)

PostPosted: Tue Apr 12, 2005 7:22 pm    Post subject: Reply with quote

This is the a table that resumes the one that fellow posted above, here we can see the thing in a compact way.
Code:

                                      ext2     ext3      jfs     xfs     reiser3  reiser4
 1   Create 10.000 files into a dir  14,950   21,473   12,144   12,756   11,278   11,642
 2   Run find on that dir            00,013   00,069   00,054   00,050   00,061   00,101
 3   Remove that dir                 00,078   00,231   00,905   01,750   00,407   00,741
 4   Create 10.000 dirs              26,667   24,928   12,001   12,863   11,098   11,870
 5   Run find on that dir            00,171   00,191   00,291   00,165   00,365   00,260
 6   Remove the whole dir            00,403   01,188   01,652   01,447   11,230   30,898
 7   Copy kernel tar from other disk 01,965   02,018   01,999   01,974   01,982   01,968
 8   Copy kernel tar to other disk   00,983   00,928   00,922   00,785   00,969   00,944
 9   Untar kernel                    34,229   36,182   36,251   37,182   36,946   38,743
10   Tar kernel                      26,639   25,859   26,824   25,806   25,942   26,569
11   Remove kerlnel source dir       00,270   00,544   04,377   01,771   01,170   02,351
12   Copy kernel tarball 10 times    19,772   20,311   18,882   19,094   20,710   17,703
13   Create 1gb from /dev/zero       23,572   25,038   26,052   22,615   25,038   23,650
14   Copy the 1gb file same disk     50,427   53,003   88,937   48,297   59,869   48,254
15   Split 10mb file 1000 byte pcs   05,769   07,939   03,477   04,262   01,390   01,140
16   Split 10mb file 1024 byte pcs   04,992   09,878   03,367   03,750   01,240   00,836
17   Split 10mb file 2048 byte pcs   01,482   02,096   01,360   01,942   00,752   00,546
18   Split 10mb file 4096 byte pcs   00,494   00,624   00,548   00,995   00,357   00,414
19   Split 10mb file 8192 byte pcs   00,229   00,278   00,347   00,510   00,257   00,310
20   Copy kernel tree same disk      08,233   19,462   23,082   19,657   27,548   14,530
21   Cat 1gb to /dev/null            22,315   22,321   22,170   22,227   24,886   22,834



My conclussions:


    :!: Managing big files, this is 9, 10, 13 and 21, makes no sensible difference, predictable, no?
    :!: Copying times are also similar (7, 8, 12)
    :!: Creation of tons of files (1, 4) is a good job for reiser3 and 4, and so bad one for ext3.
    :!: The best finding files (2, 6) are xfs and ext2, the worst are reiser brothers, and jfs in dirs.
    :!: When it comes to massive deletion (3, 6), ext2/3 is fast, reiser3 usable, the rest just sucks by long way.
    :!: To erase the kernel tree is faster with ext2/3, the worse if jfs. The rest works well also. Strange, this results seems like they contradict the previous one, someone can explain this misteriously thing? 8O


I did not look to the split marks, since, really, I dont think this is any king of usual operation, and can be greatly affected by the size of the files, the free disk space, etc, etc.

What'yaaaaa think??[/list]
_________________
Gentoo Handbook | My website
Back to top
View user's profile Send private message
Kraymer
Guru
Guru


Joined: 27 Aug 2003
Posts: 349
Location: Germany

PostPosted: Tue Apr 12, 2005 8:01 pm    Post subject: the mix Reply with quote

The mix is important.
Currently I use reiserfs (3.6) for root and home, xfs for my data partition (file server, lots of bigg files).
Once it's ready (whenever it'll be) I'll change my system and home to reiser4.

Had bad experience with jfs once: On my fileserver I had major data loss on power failure, the partition was unmounable (and believe I searched my ass off the Inet and it turned out that I was doomed (special circumstances though). Switched that to xfs, and even on a software raid, power failure happend again, system rebooted fine and everything was in its place :)

You always have to consider the purpose the filesystem shall fulfill. For overall usage, reiserfs is fine for me (although it's sad that only 8K blocksize works :( ), xfs is great for my large data partition.

Sebastian
Back to top
View user's profile Send private message
spb
Retired Dev
Retired Dev


Joined: 02 Jan 2004
Posts: 2135
Location: Cambridge, UK

PostPosted: Tue Apr 12, 2005 8:13 pm    Post subject: Reply with quote

Lechium wrote:
P.S. reiserfs4 can be the best choice for a simple reason -- a lot of people use it, so if you get into trouble it will be easier finding help, than if you run jfs/xfs.
It also means that anyone not obsessed with getting the last little bit of performance out of their system at the expense of it actually working will refuse to help you at all.

As for which to use, ext2/3 and xfs are the only ones that actually work, so I use those.
Back to top
View user's profile Send private message
Enlight
Advocate
Advocate


Joined: 28 Oct 2004
Posts: 3514
Location: Alsace (France)

PostPosted: Tue Apr 12, 2005 8:34 pm    Post subject: Reply with quote

okay so I guess I'll definitively stick with xfs, morover I tried reiserfs 3.6 once and my sata disk was making twice noise as in xfs while writing... gonna got a usb stick and try to put journalisation on it :D
Back to top
View user's profile Send private message
Kraymer
Guru
Guru


Joined: 27 Aug 2003
Posts: 349
Location: Germany

PostPosted: Tue Apr 12, 2005 9:05 pm    Post subject: sense? Reply with quote

Enlight wrote:
... gonna got a usb stick and try to put journalisation on it :D

of course you can create every filesystem you want on usb sticks... the point however is (IMO):
- to remain data 'really' portable, let it stick to fat16/32
- the media/ship/flash/whatever in usb sticks is usually really cheap. I don't know if you'll gain much be a journal if the media would fail.. :-/
Back to top
View user's profile Send private message
Enlight
Advocate
Advocate


Joined: 28 Oct 2004
Posts: 3514
Location: Alsace (France)

PostPosted: Tue Apr 12, 2005 9:08 pm    Post subject: Re: sense? Reply with quote

Kraymer wrote:
Enlight wrote:
... gonna got a usb stick and try to put journalisation on it :D

of course you can create every filesystem you want on usb sticks... the point however is (IMO):
- to remain data 'really' portable, let it stick to fat16/32
- the media/ship/flash/whatever in usb sticks is usually really cheap. I don't know if you'll gain much be a journal if the media would fail.. :-/


You didn't got it, only journalisation will be on the stick an data will be on my hard disk :

from the man:
=============

Quote:
The metadata log can be placed on another device to reduce the number
of disk seeks. To create a filesystem on the first partition on the
first SCSI disk with a 10000 block log located on the first partition
on the second SCSI disk, use:

mkfs.xfs -l logdev=/dev/sdb1,size=10000b /dev/sda1


Wonderfull...:D
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Off the Wall All times are GMT
Goto page 1, 2, 3, 4, 5  Next
Page 1 of 5

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum