Bruno Damour wrote:
Jürgen Keil wrote:
Dom0
amber ~ # uname -a
SunOS amber 5.11 snv_113 i86pc i386 i86xpv
amber ~ # /usr/gnu/bin/dd if=/dev/zero of=/tank/test count=100000
51200000 bytes (51 MB) copied, 2.70782 s, 18.9 MB/s
DomU :
corwin bruno # uname -a Linux corwin 2.6.18-xen-r12 #7 SMP Fri May
8 10:41:05 CEST 2009 x86_64 Intel(R) Core(TM)2 Quad CPU Q6600 @
2.40GHz GenuineIntel GNU/Linux
corwin bruno # dd if=/dev/zero of=/home/bruno/test count=100000
51200000 bytes (51 MB) copied, 0.504987 s, 101 MB/s
Now that is a 1 to 5 performance gap !!
Any ideas about how to explain this ?
Are you sure that linux has actually written the data to disk
when the dd has returned? In most cases (esp with such
a small amount of data) linux has the data in it's buffer
cache and *not* on the disk.
Well, that could be an explanation... I can test with more data to see
if there is any difference...
_______________________________________________
opensolaris-discuss mailing list
[email protected]
Well, I guess you were right, silly me not having thought of this
before, but I was not aware that opensolaris had a _different_ way of
doing this ? No buffering ?
Anyway, here are the new results for 2,6G :
dom0 SunOS
amber ~ # /usr/gnu/bin/dd if=/dev/zero of=test count=5000000
5000000+0 records in
5000000+0 records out
2560000000 bytes (2.6 GB) copied, 175.043 s, 14.6 MB/s
domU linux
corwin bruno # dd if=/dev/zero of=/home/bruno/test count=5000000
5000000+0 records in
5000000+0 records out
2560000000 bytes (2.6 GB) copied, 255.649 s, 10.0 MB/s
Much more logic in this...
Thanks
Bruno
_______________________________________________
opensolaris-discuss mailing list
[email protected]
