stds-1394: RE: SBP-2 Study Group

Robert Snively rsnively at Brocade.COM
Fri Aug 11 15:56:37 PDT 2000


* From the T10 Reflector (t10 at t10.org), posted by:
* Robert Snively <rsnively at Brocade.COM>
*
Peter,

I am a little bit confused about what might constitute peer to peer
behavior.  If I understand generic computer architecture, one program
wants to do something and asks another program to do it on its behalf.

SCSI is peer to peer in that any end of a link can choose to be
an initiator and ask any end to be a target.  However, the
invocation of a function is an initiator operation, and the
execution and management of the requested function is a target
operation.  If SCSI soon allows bi-directional data transfer, it is
still target managed.

So what do people mean when they say peer to peer as if it were
somehow different from SCSI?

By the way, TCP/IP has exactly the same properties.  It delivers
information, but the applications at the two ends of the TCP/IP share
authority and invocation of function capabilities that are very
much "please do this for me" and "of course I will".

Bob

>  -----Original Message-----
>  From: Peter Johansson [mailto:PJohansson at ACM.org]
>  Sent: Friday, August 11, 2000 12:17 AM
>  To: IEEE 1394
>  Cc: NCITS T10
>  Subject: Re: stds-1394: RE: SBP-2 Study Group
>  
>  
>  At 10:22 AM 8/10/00, Atsushi Nakamura wrote:
>  
>  >If it is decided that work is done on a peer to peer SBP-2, 
>  I would like 
>  >to see the outcome
>  >
>  >1) not be "yet another protocol", and it remains "SBP-2" as 
>  we see it today
>  >2) does not sacrifice the effeciency and simpleness it has 
>  today as a 
>  >master-slave transport.
>  >
>  >Though it would be ideal if SBP-2 can handle both, it would 
>  not be worth 
>  >much if too much burden (inclucing compatibility) is put on 
>  the original 
>  >protocol, or significant changes are required for the new 
>  protocol which 
>  >results in something drastically different from what we have now.
>  
>  It's hard to predict, in advance, how much interest there 
>  will be in an 
>  SBP-2 that is more peer-to-peer than the present. I think 
>  draft standard 
>  IEEE P1394.3 has done a pretty good job of enabling more symmetric 
>  connections between two SBP-2 devices, no matter which one 
>  is the initiator 
>  and which the target. With the benefit of hindsight it's 
>  easy to say that 
>  PPDT could have been a natural part of SBP-2---but now that 
>  it has been 
>  done as an add-on standard, would it be worthwhile to insert 
>  it into SBP-2?
>  
>  If there are other ideas to explore about SBP-2 and 
>  peer-to-peer, that's 
>  one thing that study group time is for.
>  
>  I understand your points, Ats, that there is a legacy base 
>  to consider 
>  whenever any modifications are made to an extant protocol. I 
>  don't think 
>  they should impede "brainstorming", else one might 
>  prematurely shut the 
>  door on new ideas. But once interesting new ideas have been 
>  created, they 
>  have to be subject to careful examination. And I think your 
>  concerns would 
>  be part of any thorough review.
>  
>  
>  
>  
>  Regards,
>  
>  Peter Johansson
>  
>  Congruent Software, Inc.
>  98 Colorado Avenue
>  Berkeley, CA  94707
>  
>  (510) 527-3926
>  (510) 527-3856 FAX
>  
>  PJohansson at ACM.org
>  
>  
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list