SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    iFCP work in the IPS WG



    After quite a bit of consultation and consideration
    of comments both from the mailing list and those
    sent directly to Steve Bellovin and myself, the
    two of us along with the Transport Area Directors
    have reached the following decision on the subject
    of working on iFCP within the IP Storage Working
    Group ...
    
    There is both sufficient technical merit in iFCP
    and interest within the ips WG in working on iFCP
    to justify adding iFCP to the official work items
    of the WG.  Hence iFCP will be added to the list of
    protocols that the WG is working on -- the following
    directions and conditions apply:
    
    - iFCP is subject to the directions, scope, constraints,
    	etc. in the IPS WG's charter, as is the case for all
    	work in the WG.  This includes requirements for
    	good transport and security designs.
    
    - FCIP and iFCP are to use a common encapsulation format, so
    	that there is one way to encode a Fibre Channel (FC-2)
    	frame for transmission over an IP network.  The header
    	must be extensible as needed by protocols that use it
    	(e.g., iFCP requires protocol-specific extensions in
    	some cases) and the header must incorporate a protocol
    	number to distinguish FCIP, iFCP, and any future protocols
    	that use it so that a recipient of an encapsulated FC
    	frame can unambiguously determine which protocol is being
    	used.
    
    - The existing WG rough consensus that iFCP and FCIP should
    	not be merged is to be respected.  The result should
    	be three related specifications - the common
    	encapsulation format, and specifications
    	for how each of iFCP and FCIP employ it.
    
    - T11.3's role as the authority on Fibre Channel and
    	Fibre Channel architecture is to be recognized
    	and respected.  Fibre Channel work in the ips WG,
    	including iFCP, will endeavor to respect and
    	follow T11.3's guidance in a fashion similar
    	to iSCSI work following T10's analogous guidance
    	on SCSI.  A specific area in which T11.3 should
    	be consulted is iFCP's modifications to the
    	S_ID and D_ID fields in Fibre Channel frames.
    
    The next steps from here are:
    - Addition of iFCP milestone(s) to the charter.  While we're
    	at it, we'll add iSNS milestone(s).  Guesstimates from
    	the various document authors would be appreciated.
    - Appointment of an iFCP technical coordinator.  Send
    	suggestions/nominations to me directly.
    - Submission of new FC encapsulation, iFCP and FCIP
    	drafts in accordance with the above.
    
    From a procedural standpoint, this decision is not
    open to debate on the mailing list, although
    clarification questions are welcome, as I continue
    to discover that I don't always write the clearest
    text despite my best efforts.
    
    Thanks,
    --David
    
    ---------------------------------------------------
    David L. Black, Senior Technologist
    EMC Corporation, 42 South St., Hopkinton, MA  01748
    +1 (508) 435-1000 x75140     FAX: +1 (508) 497-8500
    black_david@emc.com       Mobile: +1 (978) 394-7754
    ---------------------------------------------------
    
    


Home

Last updated: Tue Sep 04 01:05:31 2001
6315 messages in chronological order