FW: A Transport Protocol Without ACK (resend)

[Robert Snively <rsnively@Brocade.COM>]

-----Original Message-----
From: Robert Snively 
Sent: Thursday, September 21, 2000 8:35 AM
To: 'Douglas Otis'; Jim McGrath; 'Randall Stewart'
Cc: 'Y P Cheng'; 'Ips@Ece. Cmu. Edu'
Subject: RE: A Transport Protocol Without ACK


A minor correction:

>  
>  If you examine FCP documentation, you will find that you can 
>  send data with
>  the command as an option.  You can also send the response at 
>  the end of data
>  as an option.  Every vital feature used to justify tossing 
>  FCP structures
>  become moot.  

While this is true in FCP, the function has been made "obsolete"
in FCP-2 because there is no identifiable benefit to it and because
it complicates retry.  I believe the same would be true of any
reasonable iSCSI implementation.  A reasonable iSCSI implementation
is defined as one that has essentially zero additional overhead in 
managing two transfer units as opposed to managing one transfer unit.