gstripe small transaction size
ssanders at softhammer.net
Tue Sep 28 12:44:51 UTC 2010
Increasing MAXPHYS and turning up the stripe size won't have the
effect I'm looking for ?
I see where MAXPHYS is now a kernel configuration parameter in FreeBSD 8.1.
On 09/28/2010 06:44 AM, Ivan Voras wrote:
> On 09/28/10 03:08, Stephen Sanders wrote:
>> I'm trying a disk throughput experiment where in two 3ware raid 6's are
>> being put into a g_strip raid 0.
>> The raid 6's are using 8 7200RPM disks. The disk transfer rate is
>> ~80MB/s. Using a load generation tool that is using O_DIRECT for I/O,
>> I've generated the following short output from iostat. Needless to say,
>> the write performance is a lot less than I'm expecting.
>> We've modified the kernel such that our KB/t figure is closer to 512KB/t
>> per disk when measured without the g_strip. With g_strip turned on, the
>> KB/t number is more like 60KB/t.
>> The question is how do I get g_stripe to write larger transactions to
>> the disk ?
>> gstripe label -v -s 131072 roadkill /dev/da0 /dev/da2
> No pratical way. You have a gstripe array of two drives and with
> MAXPHYS of 128 KiB each would in fact see at most 64 KiB transactions.
> It gets worse when you add drives.
>> newfs /dev/stripe/roadkill
> Depending on what you are doing, adding "-U" would help you here.
> freebsd-performance at freebsd.org mailing list
> To unsubscribe, send any mail to
> "freebsd-performance-unsubscribe at freebsd.org"
More information about the freebsd-performance