I would normally agree, just that I get very large block sizes in the
non-NFS scenario.
However, for completeness, you'll see my standard settings allow for
pretty large transfers. The maxreceivedataseglen below is the default,
and increasing it may help, but it still doesn't answer why the local
non-nfs case has no problems, right?
First Burst Length: 65536
Immediate Data: yes
Initial Ready To Transfer (R2T): yes
Max Burst Length: 262144
Max Outstanding R2T: 1
Max Receive Data Segment Length: 8192
Max Connections: 1
Header Digest: NONE
Data Digest: NONE
On 5/6/06, Nicolas Williams <Nicolas.Williams at sun.com> wrote:
> On Fri, May 05, 2006 at 09:48:00PM -0700, Joe Little wrote:
> > Here's some sample output. Where the I write over NFS to ZFS (no
> > iscsi) I get high sizes for i/o:
> >
> > In the iscsi-backed case, I get:
>
> [small block sizes]
>
> > Looks to me the bulk of my problem is poor block size scheduling. Is
> > this tuneable for either ZFS or NFS and/or can be set?
>
> Looking at iscsiadm(1M) (and knowing little about iSCSI) I wonder what
> the maxburstlength/maxrecvdataseglen parameters are set to...
>
> Nico
> --
>
