RE: Request to exclude FC over IP from storage over IP working group charter

[somesh_gupta@hp.com]

> -----Original Message-----
> From: sob@harvard.edu [mailto:sob@harvard.edu]
> Sent: Thursday, August 10, 2000 5:38 PM
> To: somesh_gupta@hp.com
> Cc: ips@ece.cmu.edu; mankin@east.isi.edu
> Subject: Re: Request to exclude FC over IP from storage over 
> IP working
> group charter
> 
> 
> Somesh sez:
> > I would urge you to not include FC over IP in the charter of
> > the SCSI over IP working group.
> 
> there is no option for FC over IP to ignore the same set of network
> issues that the ips working group will ahev to address - that is the
> issues inherent in real-world IP networks, congestion control, 
> security, etc.  (I do not see that the security issues are all that
> different.)

I think all protocols at various levels for all sorts of applications
using IP should address congestion control, security etc. Also since
the end points are FC end points in FC over IP and they already have
protocols in place, the security issues (and
all others) and their solutions will be completely different.

> 
> It would be possible to fire up another working group to replicate 
> dealing with those issues for FC over IP  but it does seem like a
> significant waste of effort
> 
> in addition it seems to me that assuming that the only possible 
> FC over IP model is the group-of-devices-behind-a-gateway is
> short sighted - I would fully expect that vendors would quickly
> move the "gateway" into the drive so that individual drives
> could be connected and this looks quite like the rest of the work that
> ips is supposed to work on.

I think there is some misunderstanding here between what I thought was
presented by the Lucent presentor at the ietf and what I think you are
saying. The presentation at the ietf addressed was regarding an IP
tunnel between SAN islands which is what I am making all my points about.
You are talking about (I am making an assumption here so correct me if
this is not the case) a bridge here (which typically would mean storage
on IP on one-side and FCP on the other).

Again, I see all 3 as important, commercially and technically. I still
think combining FC over IP in the working group with Storage over IP
would waste more time than keeping them seperate. Keeping them seperate
would let the interested parties focus more precisely on their set of
requirements and problems and provide a relevant spec quickly.

In my opinion storage over IP is very much TCP/IP centric as it is
another important application over TCP/IP. FC over IP as discussed is
much more FC centric using IP as a tunnel. The kind of expertise
required and the people interested would be more different than the
same. I know that it will be additional burden on some senior people
in the ietf organization to track 2 working groups instead of 1. However,
I think the participants at the grunt level (and the motivations and
the expertise etc. etc) are so different that 2 is better than 1 (or so
it seems to me).

> 
> scott
> 

Do appreciate your consideration.

Somesh