lkml.org 
[lkml]   [2010]   [Feb]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: EXT4 is ~2X as slow as XFS (593MB/s vs 304MB/s) for writes?


    On Sat, 27 Feb 2010, Justin Piszcz wrote:

    >
    >
    > On Sat, 27 Feb 2010, Justin Piszcz wrote:
    >
    > >
    > >
    > > On Sat, 27 Feb 2010, Justin Piszcz wrote:
    > >
    > > >
    > > >
    > > > On Fri, 26 Feb 2010, Eric Sandeen wrote:
    > > >
    >
    > Hi,
    >
    > I have found the same results on 2 different systems:
    >
    > It seems to peak at ~350MiB/s performance on mdadm raid, whether
    > a RAID-5 or RAID-0 (two separate machines):
    >
    > The only option I found that allows it to go from:
    > 10737418240 bytes (11 GB) copied, 48.7335 s, 220 MB/s
    > to
    > 10737418240 bytes (11 GB) copied, 30.5425 s, 352 MB/s
    >
    > Is the -o nodelalloc option.
    >
    > How come it is not breaking the 350MiB/s barrier is the question?
    >
    > Justin.
    >
    >

    Besides large sequential I/O, ext4 seems to be MUCH faster than XFS when
    working with many small files.

    EXT4

    p63:/r1# sync; /usr/bin/time bash -c 'tar xf linux-2.6.33.tar; sync'
    0.18user 2.43system 0:02.86elapsed 91%CPU (0avgtext+0avgdata 5216maxresident)k
    0inputs+0outputs (0major+971minor)pagefaults 0swaps
    linux-2.6.33 linux-2.6.33.tar
    p63:/r1# sync; /usr/bin/time bash -c 'rm -rf linux-2.6.33; sync'
    0.02user 0.98system 0:01.03elapsed 97%CPU (0avgtext+0avgdata 5216maxresident)k
    0inputs+0outputs (0major+865minor)pagefaults 0swaps

    XFS

    p63:/r1# sync; /usr/bin/time bash -c 'tar xf linux-2.6.33.tar; sync'
    0.20user 2.62system 1:03.90elapsed 4%CPU (0avgtext+0avgdata 5200maxresident)k
    0inputs+0outputs (0major+970minor)pagefaults 0swaps
    p63:/r1# sync; /usr/bin/time bash -c 'rm -rf linux-2.6.33; sync'
    0.03user 2.02system 0:29.04elapsed 7%CPU (0avgtext+0avgdata 5200maxresident)k
    0inputs+0outputs (0major+864minor)pagefaults 0swaps

    So I guess that's the tradeoff, for massive I/O you should use XFS, else,
    use EXT4?

    I still would like to know however, why 350MiB/s seems to be the maximum
    performance I can get from two different md raids (that easily do 600MiB/s
    with XFS).

    Is this a performance issue within ext4 and md-raid?
    The problem does not exist with xfs and md-raid.

    Justin.




    \
     
     \ /
      Last update: 2010-02-27 12:39    [from the cache]
    ©2003-2014 Jasper Spaans. hosted at Digital Ocean