RE: RDMA over TCP (Was Re: VI (Was: Avoiding deadlock in iSCSI))

["Douglas Otis" <dotis@sanlight.net>]

Steve,

I would agree hardware acceleration would be beneficial.  SCTP would provide
significant advantages if accelerated.  With alignment maintained and out of
sequence processing allowed, the effect of accelerating SCTP would be to
allow competitive products to FC and cLAN.  TCP would only provide a
marginal improvement and never to the same levels unless you alter TCP
significantly.  Such alternation of TCP would not be beneficial however.

Doug

> -----Original Message-----
> From: owner-ips@ece.cmu.edu [mailto:owner-ips@ece.cmu.edu]On Behalf Of
> Stephen Byan
> Sent: Thursday, September 28, 2000 8:26 AM
> To: 'ips@ece.cmu.edu'
> Subject: RE: RDMA over TCP (Was Re: VI (Was: Avoiding deadlock in
> iSCSI))
>
>
> Douglas Otis [mailto:dotis@sanlight.net] wrote:
>
> > As SCSI does not benefit from VI
> > with zero copy already possible with frame alignment, the
> > area where there
> > could be commonality has been missed with the choice of CRC.
>
> I believe the point in factoring out the RDMA mechanism from iSCSI is to
> make the same hardware assists available to other application
> protocols. The
> benefit this provides to iSCSI is that it enables iSCSI adapters to be
> marketed as general-purpose application protocol accelerators, not just as
> iSCSI accelerators. This is not a technical benefit, but rather
> an economic
> and marketing benefit.
>
> Doug, do you disagree that having a generic RDMA facility would be an
> economic and marketing benefit?
>
> Regards,
> -Steve
>
> Steve Byan
> <stephen.byan@quantum.com>
> Design Engineer
> MS 1-3/E23
> 333 South Street
> Shrewsbury, MA 01545
> (508)770-3414
> fax: (508)770-2604
>