On Tue, Apr 08, 2008 at 10:09:19PM +0300, Pasi Kärkkäinen wrote:
> 
> On Mon, Apr 07, 2008 at 10:24:56PM +0300, Pasi Kärkkäinen wrote:
> > 
> > Hello list!
> > 
> > I have lvm volume (from a single sata disk) exported with IET like this:
> > 
> > Target iqn.2001-04.com.test:vol1
> >     Lun 0 Path=/dev/vg1/test1,Type=fileio
> >     InitialR2T              No
> >     ImmediateData           Yes
> > 
> > I have not modified read-ahead or other disk settings on the target server.
> > target is using deadline elevator/scheduler.
> > 
> > open-iscsi initiator (CentOS 5.1 2.6.18-53.1.14.el5PAE and the default
> > open-iscsi that comes with the distro) sees that volume as /dev/sda.
> > 
> > I'm testing performance with different read-ahead settings on the initiator.
> > 
> > Can somebody explain why I see these throughput changes? 
> > 
> 
> Do other people see the same behaviour? if not then i guess the problem (or
> a feature?) is in my target..
> 
> Or is there anything in open-iscsi that could explain this?
> 

I did some more tests..

Testing the same volume locally on the target doesn't seem to have the same
problem..

# blockdev --getra /dev/vg1/test1
256

# dd if=/dev/vg1/test1 of=/dev/null bs=1024k count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 139.841 seconds, 61.4 MB/s

# blockdev --setra 4096 /dev/vg1/test1
# blockdev --getra /dev/vg1/test1
4096

# dd if=/dev/vg1/test1 of=/dev/null bs=1024k count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 139.819 seconds, 61.4 MB/s

# blockdev --setra 1024 /dev/vg1/test1
# blockdev --getra /dev/vg1/test1
1024

# dd if=/dev/vg1/test1 of=/dev/null bs=1024k count=8192
8192+0 records in
8192+0 records out
8589934592 bytes (8.6 GB) copied, 139.89 seconds, 61.4 MB/s


Speed seems to be the same with all read-ahead values, no big changes like 
on the initiator.. (I tested with both 4k and 1024k blocksize on the target)

Then some more testing on the initiator..

read-ahead of 4096 on the initiator:

iostat on initiator:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.50    0.00    4.48   45.27    0.00   49.75

iostat on target:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.00    0.00    9.45   88.06    0.00    2.49

# dd if=/dev/sda of=/dev/null bs=4k count=524288
524288+0 records in
524288+0 records out
2147483648 bytes (2.1 GB) copied, 65.0137 seconds, 33.0 MB/s



read-ahead of 1024 on the initiator:

iostat on initiator:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.50    0.00   16.50   32.50    0.00   50.50

iostat on target:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.00    0.00   19.00   42.00    0.00   39.00

# dd if=/dev/sda of=/dev/null bs=4k count=524288
524288+0 records in
524288+0 records out
2147483648 bytes (2.1 GB) copied, 35.2285 seconds, 61.0 MB/s



read-ahead of 256 on the initiator:

iostat on initiator:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           1.00    0.00   14.43   34.83    0.00   49.75

iostat on target:
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           0.50    0.00   17.00   12.00    0.00   70.50

# dd if=/dev/sda of=/dev/null bs=4k count=524288
524288+0 records in
524288+0 records out
2147483648 bytes (2.1 GB) copied, 35.0674 seconds, 61.2 MB/s


So.. it seems with the read-ahead of 4096 (2 MB) the CPU on the 
target maxes out.. with most of the time spent on iowait. 

The question is why this happens?

Some settings I should tune on the target? 

Thanks!

-- Pasi

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to open-iscsi@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~----------~----~----~----~------~----~------~--~---

Reply via email to