Cached file read performance with 6.2-PRERELEASE

[O. Hartmann]

JoaoBR wrote:
> On Tuesday 19 December 2006 22:05, Mark Kirkwood wrote:
>> In the process of investigating performance in another area I happened
>> to be measuring sequential cached reads (in a fairly basic manner):
>>
>> $ dd if=/dev/zero of=/tmp/file bs=8k count=100000      # create file
>> 819200000 bytes transferred in 4.849394 secs (168928321 bytes/sec)

hartmann: dd if=/dev/zero of=/tmp/file bs=8k count=100000
100000+0 records in
100000+0 records out
819200000 bytes transferred in 18.540244 secs (44184964 bytes/sec)

hartmann: dd if=/dev/zero of=/tmp/file bs=8k count=100000
100000+0 records in
100000+0 records out
819200000 bytes transferred in 18.047627 secs (45391009 bytes/sec)


>>
>> $ dd of=/dev/null if=/tmp/file bs=8k                   # read it
>> 819200000 bytes transferred in 2.177922 secs (376138354 bytes/sec)

hartmann: dd of=/dev/null if=/tmp/file bs=8k
100000+0 records in
100000+0 records out
819200000 bytes transferred in 1.898039 secs (431603332 bytes/sec)

>> $ dd of=/dev/null if=/tmp/file bs=8k                   # read again
 >> 819200000 bytes transferred in 2.178407 secs (376054620 bytes/sec)

hartmann: dd of=/dev/null if=/tmp/file bs=8k
100000+0 records in
100000+0 records out
819200000 bytes transferred in 1.328635 secs (616572658 bytes/sec)

>> 819200000 bytes transferred in 2.178407 secs (376054620 bytes/sec)
>> $ dd of=/dev/null if=/tmp/file bs=32k                  # read it
>> 819200000 bytes transferred in 1.801944 secs (454620117 bytes/sec)

hartmann: dd of=/dev/null if=/tmp/file bs=32k
25000+0 records in
25000+0 records out
819200000 bytes transferred in 1.037108 secs (789888848 bytes/sec)


These are my numbers, but the box is not configured to be benchmarked 
and is a P4/3.0 GHz, HTT enabled, SMP kernel (most recent RELENG_6) with 
SCHED_ULE enabled.
HD interface is ATA100, slightly aged, I guess ...

Regards,
Oliver