Ex Parte Dibb

Board of Patent Appeals and Interferences

Dec 20, 2006

09879554 (B.P.A.I. Dec. 20, 2006)

The opinion in support of the decision being entered today was not written for
publication and is not binding precedent of the Board.

UNITED STATES PATENT AND TRADEMARK OFFICE
_______________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
_______________

Ex parte JAMES E. DIBB
_______________

Appeal No. 2006-3036
Application No. 09/879,554
_______________

ON BRIEF
_______________

Before JERRY SMITH, BARRY, and MACDONALD, Administrative Patent
Judges.

JERRY SMITH, Administrative Patent Judge.

DECISION ON APPEAL

This is a decision on the appeal under 35 U.S.C. § 134 from the
examiner’s rejection of claims 1-14 and 16-19, which constitute all the claims
pending in this application.
The disclosed invention pertains to a method and system for repairing a
redundant array of disk drives. When a disk drive fails, a mirrored subsystem is
created within the array that includes a temporary disk drive and the failed disk
drive slot. The mirrored subsystem is substituted for the failed disk drive in the
Appeal No. 2006-3036
Application No. 09/879,554

redundancy group. Each data block of the failed disk drive is reconstructed and
written to the mirrored subsystem. After a replacement disk drive is inserted into
the failed disk drive slot, data is copied from the temporary disk drive to the
replacement disk drive. The mirrored subsystem can be replaced by the
replacement disk after the data thereon matches the data on the temporary disk
drive.

Representative claim 1 is reproduced as follows:

1. A method for handling a failed disk drive in a redundancy group of disk
drives in an array of disk drives, the failed disk drive located in a failed disk drive
slot, comprising:
creating a mirrored subsystem within the array, the subsystem including a
temporary disk drive and the failed disk drive slot; and
reconfiguring the redundancy group to consist of the disk drives of the
redundancy group that have not failed and the mirrored subsystem, such that the
mirrored subsystem is substituted for the failed disk drive in the redundancy
group and the redundancy of the redundancy group is restored, when the failed
disk drive contains redundancy data for the redundancy group [sic]1

The examiner relies on the following reference:

Kedem 6,154,853 Nov. 28, 2000

The following rejections are on appeal before us:
1. Claims 1-142 stand rejected under 35 U.S.C. § 112, first paragraph, as
failing to comply with the written description requirement.

1 Claim 1 lacks a period. See brief, claims appendix. See also amendments filed Nov. 02, 2005
and Feb. 16, 2005. Although the amendment filed Feb. 16, 2005 deleted the period via a
strikethrough, we nevertheless presume that the omission of a period is a typographical error.
2 Although the examiner includes claim 15 in this rejection [see answer, page 3], claim 15 has
been cancelled. See brief, page 21 and reply brief, page 6 (indicating cancellation of claim 15).
See also answer, page 2.
2
Appeal No. 2006-3036
Application No. 09/879,554

2. Claims 16-19 stand rejected under 35 U.S.C. § 102(e) as being
anticipated by Kedem.
Rather than repeat the arguments of appellant or the examiner, we make
reference to the briefs and the answer for the respective details thereof.

OPINION
We have carefully considered the subject matter on appeal, the rejections
advanced by the examiner and the evidence of anticipation relied upon by the
examiner as support for the rejections. We have, likewise, reviewed and taken
into consideration, in reaching our decision, the appellant's arguments set forth in
the briefs along with the examiner’s rationale in support of the rejections and
arguments in rebuttal set forth in the examiner’s answer. Only those arguments
actually made by appellant have been considered in this decision. Arguments
which appellant could have made but chose not to make in the briefs have not
been considered and are deemed to be waived [see 37 CFR
§ 41.37(c)(1)(vii)(2004)].
It is our view, after consideration of the record before us, that the
specification reasonably conveys to the skilled artisan that the inventor had
possession of the invention of claims 1-14 at the time the application was filed.
Moreover, we conclude that the disclosure of Kedem fully meets the invention as
set forth in claims 16-19. Accordingly, we affirm-in-part.
3
Appeal No. 2006-3036
Application No. 09/879,554

We first consider the examiner’s rejection of claims 1-14 under 35 U.S.C.
§ 112, first paragraph as failing to comply with the written description
requirement. The examiner asserts that the specification does not reasonably
support the limitation in claim 1 calling for substituting the failed disk drive in the
redundancy group and restoring the redundancy of the redundancy group when
the failed disk drive contains redundancy data for the redundancy group [answer,
page 3]. The examiner notes that the specification teaches reconstructing each
“data block” of the failed drive [answer, pages 6 and 7]. But since the
specification distinguishes “user data” (data blocks) from “redundancy data”
(parity blocks),3 the examiner concludes that the specification refers only to
reconstructing data blocks (i.e., logical data only) – not reconstructing both data
blocks and parity blocks as claimed [answer, page 8].
Appellant argues that the application as originally filed teaches using a
mirrored subsystem to completely restore redundancy even when the failed drive
contains redundancy data. In this regard, appellant notes that the originally-filed
application expressly states that the invention reconstructs each data block of the
failed drive. Accordingly, the skilled artisan would understand that such
reconstruction includes all data sectors on the failed drive – sectors with logical
and redundancy data [brief, pages 13 and 14; reply brief, page 2; emphasis
added]. Thus, if drive A fails, the skilled artisan would understand from the
originally-filed disclosure that repairing the redundant disk drive array requires

3 Specifically, the examiner quotes the following sentence from page 5 of the specification: “Data
is stored in redundancy group 7 in stripes that contain a plurality of data blocks and at least one
associated error-correction block” [specification, page 5].
4
Appeal No. 2006-3036
Application No. 09/879,554

restoring drive A to its former condition (i.e., reconstructing all data that was
formerly on drive A). Significantly, if drive A contained logical and redundancy
data (or redundancy data only), reconstructing less than all data blocks (i.e., both
logical and redundancy data) would not restore drive A to its former condition;
consequently, the array would not be repaired. Rather, array repair under such
conditions is only possible by reconstructing both logical and redundancy data
[reply brief, pages 6 and 7].
We will not sustain the examiner’s rejection. We agree with appellant that
the skilled artisan would readily understand from the originally-filed disclosure
that reconstructing the data that was formerly on the failed drive (e.g., drive A in
the disclosure) necessarily requires reconstructing all data that was on the drive
prior to failure. Otherwise, the array would not be repaired. Therefore, if the
failed drive contains both logical and redundancy data, both the logical and
redundancy data must be restored to effect repair.
Turning to claim 1, the claim calls for restoring redundancy of the
redundancy group “when the failed disk drive contains redundancy data for the
redundancy group.” Such a limitation essentially imposes a condition: if the
failed drive should contain redundancy data for the redundancy group, then
redundancy is restored for that particular type of array by substituting the
mirrored subsystem for the failed drive.4

4 Reciting such a condition in the claim, however, raises an additional issue regarding the scope
of the claim. Specifically, we note that the claimed condition does not recite an active step, but in
effect recites alternative limitations: (1) when the failed disk drive contains redundancy data, and
(2) when the failed drive does not contain redundancy data. Significantly, dependent claims 4, 5,
5
Appeal No. 2006-3036
Application No. 09/879,554

Although individual disk drives configured in redundant arrays do not
necessarily have to contain both logical and redundancy data, the originally-filed
disclosure nevertheless teaches repairing arrays with disk drives containing both
types of data. For example, the disclosure expressly states that the redundancy
group may be a RAID-3 or RAID-5 array [specification, pages 3-5; see also
claims 4, 5, 8, and 9]. As is well known in the art, a RAID-3 array utilizes byte-
level striping with a dedicated parity drive, and a RAID-5 array utilizes block-level
data striping with distributed parity.5 In short, these arrays utilize drives that
contain redundancy data (i.e., parity data). If one of these drives should fail, all
data on the drive – including the redundancy data – must be reconstructed to
restore the drive to its former condition and effect repair in accordance with the
invention.
For at least the above reasons, the originally-filed disclosure reasonably
supports the limitations of claim 1. Accordingly, we will not sustain the
examiner’s rejection of claims 1-14 under 35 U.S.C. § 112, first paragraph.
We next consider the rejection of claims 16-19 under 35 U.S.C. § 102(e)
as being anticipated by Kedem. Anticipation is established only when a single
prior art reference discloses, expressly or under the principles of inherency, each

8, and 9 recite arrays with drives having parity (redundancy) data (RAID-3 and RAID-5 arrays).
Dependent claims 6, 7, and 10-13, however, recite arrays with drives that do not contain parity
(redundancy) data (RAID 1 and 1/0 arrays). See RAID Technology, Technick.net, at
http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=guide_raid, at 3-4 (last visited Nov.
2, 2006) (“RAID Technology”). See also id., at 7 (“Parity is used by RAID levels 2, 3, 4, and 5.
RAID 1 does not use parity because all data is completely duplicated (mirrored).”). The issue is
not before us, however, and we leave the question of the scope of the claim to the examiner and
the appellant.
5 See RAID Technology, at 3-4.
6
Appeal No. 2006-3036
Application No. 09/879,554

and every element of a claimed invention as well as disclosing structure which is
capable of performing the recited functional limitations. RCA Corp. v. Applied
Digital Data Systems, Inc., 730 F.2d 1440, 1444, 221 USPQ 385, 388 (Fed. Cir.
1984); W.L. Gore and Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1554, 220
USPQ 303, 313 (Fed. Cir. 1983).
The examiner has indicated how the claimed invention is deemed to be
fully met by the disclosure of Kedem [answer, pages 3-5]. Regarding
independent claim 16, appellant argues that Kedem does not disclose
reconstructing each data block of the failed disk drive in the redundancy group,
and writing each reconstructed data block to the mirrored subsystem as claimed
[brief, page 17; emphasis added]. Specifically, appellant contends that
“reconstructing each data block” of the failed drive as claimed must be construed
as reconstructing all data blocks – both logical and redundancy [brief, page 18].
With this construction, appellant argues that Kedem does not anticipate the claim
since Kedem does not reconstruct redundancy data of the failed drive, but rather
copies only logical data to temporary drives [id.; reply brief, page 8]. Similarly,
appellant contends that Kedem does not anticipate independent claim 18 since
Kedem does not reconstruct redundancy data blocks [brief, page 18].
We will sustain the examiner’s anticipation rejection. In short, appellant’s
arguments are not commensurate with the scope of the claims. Although
Kedem’s mirroring system does not copy the parity data stored on failing device
34 to the spare device 31, each of the failed device’s logical volumes are
7
Appeal No. 2006-3036
Application No. 09/879,554

nevertheless copied to the spare device 31 [Kedem, col. 5, lines 5-17; Figs. 4
and 5]. In our view, Kedem’s reconstructing the logical blocks of the failed drive
fully meets the limitation calling for “reconstructing each data block” giving the
term its broadest reasonable interpretation. We decline to adopt appellant’s
construction that “each data block” necessarily includes redundancy data blocks.
As we noted previously, certain RAID arrays have drives that do not contain
parity data (redundancy data), but rather contain only logical data. In short, the
scope and breadth of the term “data block” as claimed does not preclude logical
data. Accordingly, “reconstructing each data block” as claimed does not
preclude Kedem’s copying of logical data blocks.
Since Kedem fully meets independent claims 16 and 18, we will sustain
the examiner’s anticipation rejection of those claims. Since appellant has not
separately argued the patentability of dependent claims 17 and 19, these claims
fall with independent claims 16 and 18. See In re Nielson, 816 F.2d 1567, 1572,
2 USPQ2d 1525, 1528 (Fed. Cir. 1987). See also 37 CFR § 41.37(c)(vii).
In summary, we have not sustained the examiner's rejection with respect
to claims 1-14. We have, however, sustained the examiner’s rejection with
respect to claims 16-19. Therefore, the decision of the examiner rejecting claims
1-14 and 16-19 is affirmed-in-part.
8
Appeal No. 2006-3036
Application No. 09/879,554

No time period for taking any subsequent action in connection with this
appeal may be extended under 37 CFR § 1.136(a)(1)(iv).

AFFIRMED-IN-PART

JERRY SMITH )
Administrative Patent Judge )
)
)
)
) BOARD OF PATENT
LANCE LEONARD BARRY ) APPEALS AND
Administrative Patent Judge ) INTERFERENCES
)
)
)
ALLEN R. MACDONALD )
Administrative Patent Judge )

JS/jaj/rwk
9
Appeal No. 2006-3036
Application No. 09/879,554

BROMBERG & SUNSTEIN LLP
125 SUMMER STREET
BOSTON, MA 02110-1618

10