Re: New List: rdma@cisco.com: to discuss RDMA

[julian_satran@il.ibm.com]

Steph Bailey said:

I said in my previous message that using RDMA would make iSCSI
implementation easier.  I take it back.  It's not clear that this is
actually the case.  The real advantage of a general RDMA is that it
can bring the benefits traditionally associated with storage adapters
(zero copy, and low CPU overhead) to protocols other than SCSI.
However, it also seems like an advantage that a high performance iSCSI
on RDMA implementation can be built with no iSCSI specific hardware at
all.

> Let the market decide whether that one implementation should survive or
> not.  Most likely it will die as others will have been smarter in their
> designs and functional selection.

Exactly.  If the tagged transfer mode is well designed, and RDMA is
not required, nobody will do RDMA for iSCSI.

Whether to use RDMA or not is a choice that the iSCSI standard is
going to have to make.  Either way works.  However, I do not believe
there is a sensible way to allow for both possibilities.

Steph


________________

Many moths ago I was making the same arguments FOR RDMA on this list
and I faced the same strong oposition. It probably happens with every
protocol
that could benefit from a generic RDMA  but its simpler for the job at hand
to
"roll your own".  I suggest all interested parties (I include myself) to
take it off
to the separate list Costa was so kind to setup and take it on there as a
"generic
RDMA".

Julo