Re: iSCSI : Login Key issues.

[Santosh Rao <santoshr@cup.hp.com>]

Julian,

Based on Fibre Channel's FCP_XFER_RDY experiences and the WRITE I/O breakups
that occur in Fibre Channel, this value is too high. The number of
FCP_XFER_RDYs per WRITE I/O of the sizes of
256K does not require more than 4 - 5 FCP_XFER_RDYs.

We are most likely speaking ballpark figures of total no. of R2Ts for
256K WRITE I/O to be in the
range of < 10. Therefore, the number of concurrent R2Ts per task that an
initiator can handle should be in that range. It is highly unlikely that
targets are expected to fragment typical WRITE I/Os into 256 R2Ts.

Regards,
Santosh

julian_satran@il.ibm.com wrote:

> Santosh,
>
> Some of the defaults are rather arbitrary (until we gain some hands-on
> experience) but 256 is not that high for a long fat pipe. Is 255 better ?
> -:)
>
> Julo
>
begin:vcard 
n:Rao;Santosh 
tel;work:408-447-3751
x-mozilla-html:FALSE
org:Hewlett Packard, Cupertino.;SISL
adr:;;19420, Homestead Road, M\S 43LN,	;Cupertino.;CA.;95014.;USA.
version:2.1
email;internet:santoshr@cup.hp.com
title:Software Design Engineer
x-mozilla-cpt:;21088
fn:Santosh Rao
end:vcard