End-to-End Data Protection justification

Rassbach, Walter B walter.b.rassbach at intel.com
Thu Jul 3 14:23:30 PDT 2003


* From the T10 Reflector (t10 at t10.org), posted by:
* "Rassbach, Walter B" <walter.b.rassbach at intel.com>
*
This is a multi-part message in MIME format.

------_=_NextPart_001_01C341A9.59E547A8
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

In section 4.2, bullet 3, items c&d, the lengths should be 512+DIF
(i.e., 520) since the data is being received with an attached DIF (use
of DIFs was "assumed")
Similarly, in section 4.2, 7.b.iii and 8.b.iii, the length shold be
"512+DIF"
Similar corrections are needed to the diagram that follows -- The
diagram does show the DIFs being transmitted, but the length =
indications
are incorrect
=20
Other than those minor typos, I think Keith has done a very good job of
illustrating the use of the Data Integrity features in this model (in
essence, it applies to both the "simple RAID" and the "log structured"
models) and does a good job of identifying some of the kinds of errors
that may be detected and the stage at which they are caught.
=20
-- Walter Rassbach

-----Original Message-----
From: Holt, Keith [mailto:kholt at lsil.com]
Sent: Tuesday, July 01, 2003 7:52 AM
To: 'T10 Reflector'
Subject: End-to-End Data Protection justification



I mentioned in the June 23 data integrity proposal telecon that I was
preparing a document to provide some additional details regarding error
cases.  The document is now available at
ftp://ftp.t10.org/t10/document.03/03-224r0.pdf
<ftp://ftp.t10.org/t10/document.03/03-224r0.pdf>  .  I'm posting this
notice to give everyone a little extra time to review the document =
prior
to the CAP meeting since the next weekly list of uploaded documents
won't go out until a couple of days before the meeting.

Keith Holt=20

--=20
Keith Holt=20
keith.holt at lsil.com=20
LSI Logic Storage Systems=20


------_=_NextPart_001_01C341A9.59E547A8
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">

 End-to-End Data Protection justification In=20 section 4.2, bullet 3, items c&d, the lengths should be 512+DIF = (i.e., 520)=20 since the data is being received with an attached DIF (use of DIFs was=20 "assumed")
 Similarly, in section 4.2, 7.b.iii and = 8.b.iii, the=20 length shold be "512+DIF"
 Similar corrections are needed to the = diagram that=20 follows -- The diagram does show the DIFs being transmitted, but the = length=20 indications are incorrect
  
 Other=20 than those minor typos, I think Keith has done a very good job of = illustrating=20 the use of the Data Integrity features in this model (in essence, it = applies to=20 both the "simple RAID" and the "log structured" models) and does a good = job of=20 identifying some of the kinds of errors that may be detected and the = stage at=20 which they are caught.
  
 --=20 Walter Rassbach
 -----Original Message-----
From: Holt, Keith=20 [mailto:kholt at lsil.com]
Sent: Tuesday, July 01, 2003 7:52=20 AM
To: 'T10 Reflector'
Subject: End-to-End Data = Protection=20 justification


 I mentioned in the June 23 data = integrity proposal=20 telecon that I was preparing a document to provide some additional = details=20 regarding error cases.  The document is now available at = ftp://ftp.t10.org/t10/document.03/03-224r0.pdf .  I'm posting this notice to give = everyone a little=20 extra time to review the document prior to the CAP meeting since the = next=20 weekly list of uploaded documents won't go out until a couple of days = before=20 the meeting. Keith Holt -- 
Keith=20 Holt 
keith.holt at lsil.com 
LSI Logic Storage Systems=20 

------_=_NextPart_001_01C341A9.59E547A8--




More information about the T10 mailing list