Ex Parte 5819292 et al

Board of Patent Appeals and Interferences

Jul 31, 2012

90008903 (B.P.A.I. Jul. 31, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
90/008,903 12/26/2007 5819292 112056-0499C1RX 4100
24267 7590 08/01/2012
CESARI AND MCKENNA, LLP
88 BLACK FALCON AVENUE
BOSTON, MA 02210
EXAMINER
CHOI, WOO H
ART UNIT PAPER NUMBER
3992
MAIL DATE DELIVERY MODE
08/01/2012 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________________

Ex parte NETAPP, INC.,
Appellant and Patent Owner
____________________

Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292
Technology Center 3900
____________________

Before JAMESON LEE, STEPHEN C. SIU, and JUSTIN T. ARBES,
Administrative Patent Judges.

ARBES, Administrative Patent Judge.

Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

2

DECISION ON APPEAL
NetApp, Inc., the owner of Patent 5,819,292, the patent under
reexamination (hereinafter the “‘292 Patent”), appeals under 35 U.S.C. §§
134(b) and 306 from a final rejection of claims 4, 8, 11-15, 20, 23, and 24.
Claims 1, 21, and 22 were confirmed. Claims 2, 3, 5-7, 9, 10, and 16-19
were not subject to reexamination. We have jurisdiction under 35 U.S.C. §§
134(b) and 306.1
We affirm.

STATEMENT OF THE CASE
This reexamination proceeding arose from two third-party requests for
ex parte reexamination filed by Ronald L. Yin of DLA Piper US LLP: (1) a
Corrected Request for Ex Parte Reexamination filed December 26, 2007, the
subject of reexamination control number 90/008,903, and (2) a Request for
Ex Parte Reexamination filed April 4, 2008, the subject of reexamination
control number 90/009,103. The two reexamination proceedings were
merged. See Decision Sua Sponte Merging Reexamination Proceedings,
September 5, 2008, at 3. The merged proceeding subsequently was
remanded to the Examiner for consideration of Appellant’s petition to
designate the Examiner’s Answer as containing a new ground of rejection.
See Order Remanding Appeal to Examiner, July 27, 2010. Appellant’s

1 Our decision will make reference to Appellant’s Appeal Brief (“App. Br.,”
filed February 16, 2010) and Reply Brief (“Reply Br.,” filed June 21, 2010),
and the Examiner’s Answer (“Ans.,” mailed April 19, 2010) and Final
Rejection (“Final Rej.,” mailed June 16, 2009).
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

3

petition was denied, making this appeal now ripe for review. See Decision
Denying Petition Under 1.181, January 18, 2012.
Appellant states that this appeal is related to the following other
reexaminations (App. Br. 6):
Reexamination Control No. 95/000,328 (Appeal 2012-009493)
involving Patent 6,892,211 B2, which claims priority to the
‘292 Patent;
Reexamination Control No. 90/009,129 (Appeal 2012-003428)
involving Patent 7,174,352 B2, which claims priority to the
‘292 Patent;
Reexamination Control No. 95/000,324 (Appeal 2012-002709)
involving Patent 6,857,001 B2, which incorporates the ‘292
Patent by reference; and
Reexamination Control No. 95/000,373 involving Patent
7,313,720 B1, which incorporates the ‘292 Patent by reference.
The ‘292 Patent relates to methods of maintaining a consistent file
system and for creating read-only copies of a file system (col. 4, ll. 7-9).
Claim 4 is exemplary of the claims on appeal:
4. A method for maintaining a file system comprising
blocks of data stored in blocks of a non-volatile storage means
at successive consistency points comprising the steps of:
storing a first file system information structure for a first
consistency point in said non-volatile storage means, said first
file system information structure comprising data describing a
layout of said file system at said first consistency point of said
file system;
writing blocks of data of said file system that have been
modified from said first consistency point as of the
commencement of a second consistency point to free blocks of
said non-volatile storage means;
storing in said non-volatile storage means a second file
system information structure for said second consistency point,
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

4

said second file system information structure comprising data
describing a layout said file system at said second consistency
point of said file system.
Claims App’x, App. Br. 29.
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Mendel Rosenblum & John K. Ousterhout, The Design and
Implementation of a Log-Structured File System, Proceedings
of the 13th ACM Symposium on Operating Systems Principles,
1991, pp. 1-15 (“Rosenblum”);
Bob Sidebotham, Volumes: The Andrew File System Data
Structuring Primitive, EUUG Conference Proceedings,
Manchester, UK, Autumn 1986, pp. 1-8 (“Sidebotham”);
Sean Quinlan, A Cached WORM File System, Software –
Practice and Experience, Vol. 21 (12), Dec. 1991, pp. 1289-99
(“Quinlan”); and
Jason Gait, Phoenix: A Safe In-Memory File System,
Communications of the ACM, Vol. 33, No. 1, Jan. 1990, pp.
81-86 (“Gait”).
The Examiner (1) rejected claim 4 under 35 U.S.C. § 102(b) as being
anticipated by Rosenblum; (2) rejected claim 4 under 35 U.S.C. § 102(b) as
being anticipated by Sidebotham; (3) rejected claims 4, 8, 11-15, and 20
under 35 U.S.C. § 102(b) as being anticipated by Quinlan; and (4) rejected
claims 4, 8, 11-15, 20, 23, and 24 under 35 U.S.C. § 102(b) as being
anticipated by Gait. Ans. 3-13.

ISSUES
The principal issues in this appeal are: (1) did the Examiner err in
interpreting the terms “consistency point,” “file system information
structure,” and “marking,” and (2) did the Examiner incorrectly determine
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

5

that claims 4, 8, 11-15, 20, 23, and 24, when properly interpreted, are
anticipated by each of Rosenblum, Sidebotham, Quinlan, and Gait?

PRINCIPLES OF LAW
“During reexamination, as with original examination, the PTO must
give claims their broadest reasonable construction consistent with the
specification. . . . Therefore, we look to the specification to see if it provides
a definition for claim terms, but otherwise apply a broad interpretation.” In
re ICON Health and Fitness, Inc., 496 F.3d 1374, 1379 (Fed. Cir. 2007).
We must be careful not to read a particular embodiment appearing in the
written description into the claim if the claim language is broader than the
embodiment. See In re Van Geuns, 988 F.2d 1181, 1184 (Fed. Cir. 1993)
(“[L]imitations are not to be read into the claims from the specification.”).
“Although an inventor is indeed free to define the specific terms used
to describe his or her invention, this must be done with reasonable clarity,
deliberateness, and precision. ‘Where an inventor chooses to be his own
lexicographer and to give terms uncommon meanings, he must set out his
uncommon definition in some manner within the patent disclosure’ so as to
give one of ordinary skill in the art notice of the change.” In re Paulsen, 30
F.3d 1475, 1480 (Fed. Cir. 1994) (citation omitted).
“Anticipation requires that every limitation of the claim in issue be
disclosed, either expressly or under principles of inherency, in a single prior
art reference.” Corning Glass Works v. Sumitomo Elec. U.S.A., Inc., 868
F.2d 1251, 1255-56 (Fed. Cir. 1989).

Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

6

ANALYSIS
I. Claim Interpretation
We begin by interpreting the claim terms “consistency point” (claims
4 and 20), “file system information structure” (claim 4), and “marking”
(claims 8, 23, and 24), as Appellant argues that the Examiner erred in
interpreting each term. See App. Br. 14-18; Reply Br. 9-19.

A. “Consistency Point” (Claims 4 and 20)
Appellant argues that “consistency point” is explicitly defined in the
‘292 Patent to mean a “set of self-consistent blocks on disk that is rooted by
a root inode,” and that this interpretation is consistent with the surrounding
language of the claims and how the term is used throughout the
Specification. App. Br. 14-15 (citing ‘292 Patent, col. 4, ll. 11-13, and col.
11, l. 66-col. 12, l. 1); Reply Br. 9-15 & n. 2. Appellant also points to
dependent claim 20, which recites “using one or more of said read-only
copies of said file system to back-up said blocks comprising one or more
consistency points of said file system,” as support for its view that a
“consistency point” is a set of blocks on disk. Reply Br. 12-13.
The Examiner agrees that the ‘292 Patent uses “consistency point” in
the manner proposed by Appellant at certain points in the Specification, but
finds that the Specification also uses the term differently in other situations.
Ans. 13-14. In particular, the Specification states that “[a] new consistency
point occurs when the fsinfo block is updated,” and the generation count of
the fsinfo block is updated “[e]ach time a consistency point (or snapshot) is
performed.” Ans. 13-14 (citing ‘292 Patent, col. 4, ll. 15-17, and col. 14, ll.
35-37) (emphases added). The Examiner therefore determined that based on
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

7

the Specification, “consistency point” can mean either (1) a set of self-
consistent blocks on disk that is rooted by a root inode, (2) a point at which
the file system is consistent, or (3) an act that can be performed, depending
on the context in which the term is used. Ans. 13-14, 17. The Examiner
determined that the second interpretation, which is also the ordinary
meaning of the term, was the most reasonable in the context of claims 4 and
20. Ans. 17.
We agree with the Examiner that the term “consistency point” in
claims 4 and 20 encompasses a point at which the file system is consistent.
Appellant is correct that the Specification provides: “The set of self-
consistent blocks on disk that is rooted by the root inode is referred to as a
consistency point (CP).” ‘292 Patent, col. 4, ll. 11-13. In the absence of any
other evidence, such a statement might indicate that the inventors intended
to be their own lexicographer. But that is not the case here. The
Specification also uses the term more broadly in stating that a consistency
point “occurs when” a file system information structure is updated and that a
consistency point is “performed” at a certain time. ‘292 Patent, col. 4, ll. 15-
17; col. 14, ll. 35-37; see Ans. 13-14. The Specification further provides
that the state of the file system is flushed to disk exactly as it was “when the
consistency point began,” and that a consistency point is “completed” in step
550 of Figure 5. ‘292 Patent, col. 12, ll. 57-58; col. 14, ll. 11-13; Reply Br.
10 (acknowledging that file system information structures are stored “to
complete” the first and second consistency points in claim 4). A person of
ordinary skill in the art would not understand a set of blocks on disk to
“begin,” “occur,” “complete,” or be “performed.” Thus, while Appellant is
correct that the inventors used “consistency point” at some points in the
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

8

Specification to refer to a particular set of blocks stored on disk, the
Specification also indicates that the inventors contemplated different
meanings for the term depending on the context in which the term is used.
Accordingly, we are not persuaded by Appellant’s argument that the
Specification explicitly defines “consistency point” in all circumstances to
be a set of self-consistent blocks on disk that is rooted by a root inode. See
Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985, 991 (Fed.
Cir. 1999) (“Varied use of a disputed term in the written description
demonstrates the breadth of the term rather than providing a limited
definition.”).
Turning to the claims themselves, the second potential interpretation
of “consistency point” is consistent with the surrounding language of the
claims. See Phillips v. AWH Corp., 415 F.3d 1303, 1314 (Fed. Cir. 2005)
(“the context in which a term is used in the asserted claim can be highly
instructive”). Claim 4 recites writing modified blocks of data “as of the
commencement of” a second consistency point. Appellant’s proposed
interpretation of a set of blocks on disk does not make sense if plugged in to
this limitation of claim 4, as it is not clear how modified blocks of data can
be written “as of the commencement of” a set of blocks on disk. See Ans.
17. Stated another way, claim 4 does not recite writing a consistency point;
it recites writing modified blocks of data as of the commencement of a
second consistency point. Appellant argues that “it is clear to a person of
ordinary skill in the art that writing of the file system blocks that have been
modified from the first consistency point to free blocks of the disk
represents the start (commencement) of the formation of a second
consistency point.” Reply Br. 15 (emphasis added). Claim 4, however, does
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

9

not include the word “represents.” Thus, Appellant’s argument contradicts
the express language of claim 4.
The Examiner’s interpretation also makes sense in the context of the
other limitations of claim 4, which recite storing file system information
structures “for” first and second consistency points where the structures
comprise data describing a layout of the file system “at” the consistency
points.
Further, given the term’s broad usage in the Specification and the
language of claim 4 discussed above, we cannot say that the Examiner’s
interpretation is unreasonable in the context of dependent claim 20 either.
Claim 20 recites storing meta-data identifying blocks of a non-volatile
storage means for successive “states” of a file system, making copies of the
meta-data, marking the identified blocks for each copy as comprising a read-
only copy of the file system, and “using one or more of said read-only copies
of said file system to back-up said blocks comprising one or more
consistency points of said file system.” The point at which the file system is
consistent corresponds to the point at which meta-data identifying blocks at
a particular “state” is stored and copied; thus, it is more natural to say that a
particular set of blocks “comprises” a consistency point than to say that a
consistency point is the set of blocks as Appellant suggests. Appellant’s
interpretation also would result in a partial redundancy in the claim where
“blocks” would “compris[e]” a set of blocks. Faced with the two proposed
interpretations for “consistency point” as it appears in claims 4 and 20, we
conclude that the Examiner’s interpretation is the most reasonable.
Appellant has not pointed to persuasive evidence showing why the
Examiner’s interpretation of “consistency point” is unreasonable given the
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

10

term’s broad usage in the Specification and in the claims themselves.
Applying the broadest reasonable interpretation of the claim in light of the
Specification, we interpret the term “consistency point” to encompass a
point at which the file system is consistent.

B. “File System Information Structure” (Claim 4)
Appellant makes two arguments with respect to the term “file system
information structure” in claim 4. First, Appellant argues that the term
means “a data structure that includes a root inode and may include additional
file system information.” App. Br. 16-17; Reply Br. 16-18. Appellant
contends that the root inode is a “necessary part” of the file system
information structure that “must always be present in that structure,” citing
Figure 15 and the accompanying description in the Specification. Reply Br.
16-17 (citing ‘292 Patent, col. 10, l. 57-col. 11, l. 5). The Examiner,
however, interpreted the term to mean a data structure that contains file
system information. Ans. 16.
We do not find support in claim 4 or the portions of the Specification
cited by Appellant for a requirement that the “file system information
structure” of claim 4 always include a root inode. Claim 4 recites that the
first and second file system information structures each comprise “data
describing a layout of said file system at” a consistency point, but does not
mention anything regarding inodes or root inodes (unlike original claim 19,
for instance, which depends from claim 8 and recites a “root inode”). Figure
15 also illustrates “a file system information (fsinfo) structure 1510,” and
does not explicitly define “file system information structure” or otherwise
indicate that fsinfo structure 1510 (with all of its attendant features) is itself
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

11

the “file system information structure” of claim 4 rather than merely an
exemplary embodiment. See ‘292 Patent, col. 10, ll. 57-58 (emphasis
added); Resonate Inc. v. Alteon Websystems, Inc., 338 F.3d 1360, 1367 (Fed.
Cir. 2003) (“limitations may not be read into a claim from a preferred
embodiment when the claim language is broader than that embodiment”).
Finally, to the extent Appellant’s argument regarding the root inode relies on
its interpretation of “consistency point,” we disagree for the reasons stated
above. Applying the broadest reasonable interpretation of the term in light
of the Specification, as we must, we agree with the Examiner that “file
system information structure” encompasses a data structure that contains file
system information. See Ans. 14-16.
Second, Appellant argues that when “consistency point” and “file
system information structure” are read together, the file system of claim 4
must always be in a consistent state. App. Br. 17-20; Reply Br. 11-12. As
support, Appellant cites column 12, lines 9-24 of the Specification, which
states in part: “With this method for atomically updating a file system, the
on-disk file system is never inconsistent.” See Reply Br. 11-12; ‘292 Patent,
col. 12, ll. 14-16.
Again, we see nothing in claim 4 or the Specification indicating that
the “never inconsistent” file system described with respect to Figure 16 is a
requirement of the claim rather than an exemplary embodiment. See ‘292
Patent, col. 12, ll. 14-16. Claim 4 recites in broad terms storing first and
second file system information structures for first and second consistency
points; it does not, for example, speak to what may happen before or after
such points.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

12

Also, the Specification does not indicate that the disclosed file system
can never be inconsistent per se, but rather that the disclosed system remains
consistent by virtue of the data blocks for one consistency point (an old point
or a new point) always being stored on disk:
Referring to FIG. 16, a new consistency point is written by first
flushing all file system blocks to new locations on disk
(including the blocks in meta-data files such as the inode file
1620, blkmap file 1630, and inomap file 1640). A new root
inode 1610B and 1612B for the file system 1670 is then written
to disk. With this method for atomically updating a file system,
the on-disk file system is never inconsistent. The on-disk file
system 1670 reflects an old consistency point up until the root
inode 1610B and 1612B is written. Immediately after the root
inode 1610B and 1612B is written to disk, the file system 1670
reflects a new consistency point.
‘292 Patent, col. 12, ll. 9-20 (emphasis added); see also ‘292 Patent, col. 13,
ll. 15-18 (step 530 in Figure 5 is “the only point in time when the file
system, except for the fsinfo block, is completely self consistent and about to
be written to disk”); col. 13, l. 63-col. 14, l. 10. In the disclosed
embodiment, (1) a root inode is written atomically to disk, indicative of the
first consistency point, (2) modified data blocks are flushed from memory to
new locations on disk, and (3) a new root inode is written atomically to disk,
indicative of the second consistency point. Should a system crash occur in
the middle of the second step, the file system on disk may be considered
inconsistent overall as only some of the necessary data had been written, but
would still remain in the first consistency point until the new root inode is
written. See Reply Br. 23 n.4 (“the on-disk file system would remain
consistent as reflected by the first (old) consistency point”). Thus, the
Specification does not support a consistency requirement for claim 4 other
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

13

than what is expressly recited in the claim: storing a first file system
information structure for a first consistency point, and later writing modified
data blocks and storing a second file system information structure for a
second consistency point.
Accordingly, we agree with the Examiner’s interpretation and do not
interpret claim 4 as requiring the file system to always be in a consistent
state. See Ans. 18-20.2

C. “Marking” (Claims 8, 23, and 24)
Appellant argues that the term “marking” in claims 8, 23, and 24
requires “the ability to be reversed.” App. Br. 17-18; Reply Br. 18-19.
Appellant contends that dependent claim 15, which recites the method of
claim 8 further comprising the step of “unmarking” the marked blocks,
defines the term “marking” by implication to include reversibility. App. Br.
18. According to Appellant, “‘marking’ cannot have one meaning (e.g.,
non-reversible), while reading that limitation in claims 8, 23, and 24, and a
different meaning (e.g., reversible) when reading that limitation as part of
claim 15).” App. Br. 18; Reply Br. 19. We disagree with Appellant.
Appellant’s interpretation improperly reads a limitation from
dependent claim 15 into independent claims 8, 23, and 24. See Ans. 22.
“Unmarking” marked blocks is an additional feature recited by dependent

2 Appellant similarly argues that the Examiner’s interpretation of
“consistency point” as a point at which the file system is consistent is
“essentially meaningless and therefore unreasonably broad” because the file
Footnote continued on next page.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

14

claim 15 and is not required by the independent claims, which are
presumptively of different scope. See Karlin Tech., Inc. v. Surgical
Dynamics, Inc., 177 F.3d 968, 971-72 (Fed. Cir. 1999) (claim differentiation
“normally means that limitations stated in dependent claims are not to be
read into the independent claim from which they depend”). Using the
Examiner and Appellant’s analogy (Ans. 22; Reply Br. 27), a pen and pencil
may both be capable of marking and therefore meet a “marking”
requirement of an independent claim, but only the pencil would be capable
of meeting an additional “unmarking” requirement of a dependent claim.
Reading in a reversibility requirement to independent claims 8, 23, and 24
therefore would be improper. See Wenger Mfg., Inc. v. Coating Mach. Sys.,
Inc., 239 F.3d 1225, 1234 (Fed. Cir. 2001) (“the ‘air circulation means’
limitation in claim 1 should be limited to structure for performing the recited
function of circulating air, and should not be interpreted as requiring
structure capable of performing the additional function of recirculation,
which is expressly recited in dependent claim 3 and not found in claim 1”)
(emphasis added).
Dependent claim 15 does not define “marking” by implication, and
interpreting independent claims 8, 23, and 24 as not requiring reversibility
does not render the claims inconsistent with claim 15. Accordingly, we do
not interpret “marking” as requiring the ability to be reversed.

system must always be consistent. Reply Br. 11-12. We disagree that the
file system must always be consistent, for the reasons discussed above.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

15

II. Anticipation Rejections
Having interpreted the claim limitations disputed by Appellant, we
turn to the substance of the Examiner’s anticipation rejections.

A. Rejection Based on Rosenblum (Claim 4)
Appellant’s arguments with respect to Rosenblum are premised on its
proposed interpretations for “consistency point” and “file system
information structure.” Appellant argues that Rosenblum does not disclose
the first and second “consistency point” limitations of claim 4 because
Rosenblum fails to disclose (1) a “root inode,” which is allegedly part of the
definition of “consistency point,” and (2) a file system that is always in a
consistent state, which is allegedly required by the combination of
“consistency point” and “file system information structure” in the claim.
App. Br. 18-20; Reply Br. 20-24. In light of the claim interpretations
discussed above, we are not persuaded by Appellant’s arguments. As the
Examiner found, Rosenblum discloses storing a first file system information
structure (data contained in a first checkpoint written to the log) comprising
data describing the file system layout at a first point at which the file system
is consistent, writing modified blocks of data, and storing a second file
system information structure (data contained in a second checkpoint)
comprising data describing the file system layout at a second point at which
the file system is consistent. See Ans. 4, 16-20; Rosenblum § 4.1 (the
system “writes out all modified information to the log, including file data
blocks, indirect blocks, inodes, and blocks of the inode map and segment
usage table”). Accordingly, we sustain the rejection of claim 4 as
anticipated by Rosenblum.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

16

B. Rejection Based on Sidebotham (Claim 4)
Similar to the rejection based on Rosenblum, Appellant’s arguments
with respect to Sidebotham are premised on its proposed claim
interpretations. Appellant argues that Sidebotham does not disclose (1) a
“root inode” that can be updated, and (2) a file system that is always in a
consistent state. App. Br. 20-21; Reply Br. 24-25. In light of the claim
interpretations discussed above, we are not persuaded by Appellant’s
arguments. As the Examiner found, Sidebotham discloses storing a first file
system information structure (data contained in a first clone read-only copy)
comprising data describing the file system layout at a first point at which the
file system is consistent, writing modified blocks of data, and storing a
second file system information structure (data contained in a second clone)
comprising data describing the file system layout at a second point at which
the file system is consistent. See Ans. 4-5, 20-21; Sidebotham, pp. 5-6
(“The clone operation makes an identical, read-only copy of a volume by
copying the header and vnode index and incrementing the count on all of the
inodes referenced by the index.”). Accordingly, we sustain the rejection of
claim 4 as anticipated by Sidebotham.

C. Rejection Based on Quinlan (Claims 4, 8, 11-15, and 20)
Claims 4 and 20
Similar to the rejections based on Rosenblum and Sidebotham,
Appellant’s arguments with respect to Quinlan and claims 4 and 20 are
premised on its proposed claim interpretations. Appellant argues that
Quinlan does not disclose (1) a “root inode” that can be updated, and (2) a
file system that is always in a consistent state. App. Br. 21-22; Reply Br.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

17

25-27. In light of the claim interpretations discussed above, we are not
persuaded by Appellant’s arguments. As the Examiner found and as shown
in Figures 3a-b, Quinlan discloses storing a first file system information
structure (data contained in a first snapshot) comprising data describing the
file system layout at a first point at which the file system is consistent,
writing modified blocks of data, and storing a second file system
information structure (data contained in a second snapshot) comprising data
describing the file system layout at a second point at which the file system is
consistent as recited in claim 4. See Ans. 6-7, 21; Quinlan, pp. 1290 (“At
the user’s request, the file system freezes activity and flushes the WORM
[write-once-read-many] cache, thus ensuring that all data is on the WORM.
A consistent snapshot of the file system appears within seconds in a parallel
read-only file system. . . .”), 1294 (“A pointer to the root block is stored as a
new entry in the root directory of the dump file system.”). And for similar
reasons, we agree with the Examiner’s finding that Quinlan discloses using a
read-only copy of the file system to back-up blocks comprising a point at
which the file system is consistent as recited in claim 20. See Ans. 9, 21;
Quinlan, pg. 1292 (“One of the main goals of this file system is to provide a
convenient and efficient means of recording snapshots of the entire file
system. These snapshots reside entirely on the WORM, which is both
reliable and write-once. . . .”). Accordingly, we sustain the anticipation
rejection of claims 4 and 20.3

3 Claim 20 depends from claim 8. We do not find Appellant’s additional
arguments regarding claim 8 persuasive, as discussed below. Thus, we
sustain the rejection of claim 20 as anticipated by Quinlan.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

18

Claims 8 and 11-14
Claim 8 recites “marking said blocks of said non-volatile storage
means identified in said meta-data as comprising a respective read-only copy
of said file system,” and claim 15 recites “unmarking said blocks . . . when
said read only copy of said file system is no longer needed.” Appellant’s
sole argument with respect to claim 8 is that the marked blocks copied onto
the WORM drive in Quinlan are incapable of being unmarked. App. Br. 22-
23; Reply Br. 27-29. Because we do not interpret the term “marking” to
require reversibility, for the reasons discussed above, we are not persuaded
by Appellant’s argument. We agree with the Examiner’s finding that
Quinlan discloses “marking” blocks as read-only as recited in claim 8, and
therefore sustain the anticipation rejection of claim 8, as well as dependent
claims 11-14, which are not separately argued. See Ans. 7, 22; Quinlan, pp.
1290-91 (dump and read blocks are “read-only”); App. Br. 23 (“all blocks
that are copied to Quinlan’s WORM drive are permanently ‘marked’ read-
only”).

Claim 15
Appellant argues with respect to claim 15 that Quinlan does not
disclose “unmarking” blocks marked as read-only. App. Br. 23-24; Reply
Br. 27-30. Appellant contends that that there is no express teaching in
Quinlan that read-only blocks on the WORM cache (i.e., blocks marked as
dump or read) are ever reclaimed like write blocks. App. Br. 23. Appellant
points to Figure 1 of Quinlan, which shows a transition path from write to
free but not from dump or read back to either of the writable states:
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

19

Figure 1 of Quinlan depicts the state transition diagram for a block.
The Examiner’s position is that while Quinlan does not expressly
disclose reclaiming read-only blocks, a person of ordinary skill in the art
would understand the reference’s statement that “[a] block can return to the
free state if the block is subsequently not needed, for example when a file is
deleted,” to mean reclaiming read-only blocks for a deleted file. Ans. 23
(citing Quinlan, pg. 1290). The Examiner further finds that if Quinlan did
not reclaim read-only blocks, “the WORM cache will cease to function very
quickly because using the dump command converts all write blocks into read
only blocks.” Ans. 23.
Based on the record before us, we find Appellant’s argument
persuasive. Figure 1 of Quinlan does not show any return transitions from
the read-only states, and Quinlan’s discussion of reclaiming “not needed”
blocks is in the context of discussing transitions to and from the write state,
not transitions to and from the read-only states. Quinlan, pp. 1290-91, 1296
(“When a file is deleted, all write blocks in the file are marked free and
added to the free list.”). Further, as Appellant points out, Quinlan discloses
increasing the number of free blocks on the WORM cache “[w]hen the free
list becomes empty,” which would not be necessary if the read-only blocks
in the cache could be reclaimed. See Reply Br. 29-30 (citing Quinlan, pg.
1296). Accordingly, we do not sustain the rejection of claim 15 as
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

20

anticipated by Quinlan.

D. Rejection Based on Gait (Claims 4, 8, 11-15, 20, 23, and 24)
Claims 4 and 20
Similar to the rejections based on Rosenblum, Sidebotham, and
Quinlan, Appellant argues that based on its proposed claim interpretations,
Gait does not disclose (1) a “root inode” that can be updated, and (2) a file
system that is always in a consistent state. App. Br. 24-25; Reply Br. 30-33.
In light of the claim interpretations discussed above, we are not persuaded
by Appellant’s arguments. As the Examiner found and as shown in Figure
3, Gait discloses storing a first file system information structure (data
contained in a first timestamped version of the file system) comprising data
describing the file system layout at a first point at which the file system is
consistent, writing modified blocks of data, and storing a second file system
information structure (data contained in a second timestamped version)
comprising data describing the file system layout at a second point at which
the file system is consistent as recited in claim 4. See Ans. 9-10, 24-25;
Gait, pp. 81 (“Phoenix starts with a timestamped and write-protected version
of the file system and evolves it through a copy on write strategy while
leaving the original version undisturbed. The new version of the file system
is eventually timestamped. . . .”), 82 (“The contents of the root node are
pointers to memory locations where the interior nodes are found. . . .”). And
for similar reasons, we agree with the Examiner’s finding that Gait discloses
using a read-only copy of the file system to back-up blocks comprising a
point at which the file system is consistent as recited in claim 20. See Ans.
11-12, 24-25; Gait, pp. 84 (“The reserved timestamp preserves a consistent
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

21

logical disk in the event of a failure.”), 83 (“A lock is provided so garbage
collection can be delayed while a consistent version of the file system is
recorded on a hard disk, even when the system is not quiescent.”).
Appellant also argues that Gait does not disclose a “file system” as
required by claims 4 and 20. App. Br. 25; Reply Br. 31-32. Appellant
contends that a person of ordinary skill in the art would understand a “file
system” to be a system for managing files, and Gait’s logical disk system
does not have any data structures or mechanisms to perform such file
management. App. Br. 25. According to Appellant, “an actual file system
(such as the Unix file system) would overlay the logical disk of Gait” and
provide the necessary data structures for managing files. App. Br. 25; Reply
Br. 31-32. Appellant’s argument is not persuasive.
The title of Gait is “Phoenix: A Safe In-Memory File System,” and
the reference repeatedly refers to the Phoenix system as a “file system.” See,
e.g., Gait, pp. 81, 83-84. Gait further discloses a tree structure with a root
node and a set of blocks rooted by the root node for storing file data, as the
Examiner found. See Ans. 25; Gait, pp. 83-84. We also note that
Appellant’s argument regarding Gait was found unpersuasive for a similar
claim reciting a “file system” in related Appeal 2011-003428, Ex parte
NetApp, Inc., 2011 WL 1557901, at *8 (BPAI Apr. 22, 2011). We disagree
with Appellant for similar reasons. Accordingly, we sustain the anticipation
rejection of claims 4 and 20.4

4 Claim 20 depends from claim 8. We do not find Appellant’s additional
arguments regarding claim 8 persuasive, as discussed below. Thus, we
sustain the rejection of claim 20 as anticipated by Gait.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

22

Claims 8, 11-14, 23, and 24
Appellant argues with respect to the “marking” limitations of claims
8, 23, and 24 that the blocks marked as read-only in Gait are incapable of
being unmarked. App. Br. 26-27. Because we do not interpret the term
“marking” to require reversibility, for the reasons discussed above, we are
not persuaded by Appellant’s argument. We likewise are not persuaded by
Appellant’s argument that Gait fails to disclose a “file system” as required
by claims 8, 23, and 24, for the reasons discussed above. See App. Br. 27;
Reply Br. 33.
Appellant also argues that Gait does not disclose making a “copy” of
meta-data for a state of the file system and then “marking” blocks identified
in the meta-data copy as read-only. App. Br. 26-27; Reply Br. 33.
Appellant contends that the bottom file system state shown in Figure 3 of
Gait is not a “copy” of the top state because “they are different structures
containing different data,” and even if the bottom state was a “copy,” Gait
marks the top state (the previous state) as read-only, not the bottom state (the
copy). App. Br. 26-27; Reply Br. 33. We disagree with Appellant.
Gait is directed to a file system that protects itself in the event of a
crash by always keeping a “consistent version” of the file system in battery-
powered memory. Gait, pp. 81, 83. It does so through a repetitive copy-on-
write process of writing timestamped versions of the file system and
reclaiming (garbage collecting) memory for the “old” version once a “new”
timestamp is written:
Phoenix starts with a timestamped and write-protected version
of the file system and evolves it through a copy on write
strategy while leaving the original version undisturbed. The
new version of the file system is eventually timestamped: the
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

23

replaced pages in the old version are garbage collected; and the
process begins again with the newly timestamped version used
as the reserve file system.
Gait, pg. 81.

Figure 3 of Gait depicts the two timestamps: one for the “reserved file
system” that is “write protected”/read-only and one for the “evolving” file
system.
We agree with the Examiner that Gait discloses periodically making a
“copy” of stored meta-data for file system states. Ans. 10, 12-13, 26 (citing
Gait, Fig. 3). As shown in Figure 3, one of the interior nodes for the “new”
timestamped version points to the same leaf node as the corresponding
interior node for the “old” version, indicating that the corresponding data has
not been modified. Gait, pg. 84 (“unmodified leaf nodes are not garbage
collected as they are still pointed to by their interior node entries in the new
version of the file system”). Thus, at least the pointer for that node is a copy
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

24

of the original pointer. Indeed, if no changes are made to the file system
from one periodic timestamp to the next, all of the “new” interior node
pointers would be copies of the “old” versions. See Gait, pg. 83 (“the
frequency of timestamping can be high”). Further, as the Examiner found,
while the bottom state shown in Figure 3 is not marked as read-only when
first created, it is marked as read-only when the next timestamp is written
(i.e., when it becomes the “old” timestamp). See Ans. 26; Gait, pp. 81, 83.
We do not see any requirement in claims 8, 23, and 24 that the blocks be
marked as read-only at the same time the copy is made, and therefore agree
with the Examiner that Gait discloses the “marking” limitations of the
claims. Accordingly, we sustain the anticipation rejection of claims 8, 23,
and 24, as well as dependent claims 11-14, which are not separately argued.

Claim 15
Appellant argues with respect to claim 15 that Gait does not disclose
“unmarking” blocks marked as a read-only “copy” because it is the top state
(the previous state) that is marked as read-only and then garbage collected,
not the bottom state (the copy). App. Br. 27-28; Reply Br. 34. We do not
find Appellant’s argument persuasive for reasons similar to those discussed
above for claims 8, 23, and 24. As the Examiner found, the bottom state
shown in Figure 3 of Gait is marked as read-only when the next timestamp is
written, and is then garbage collected when no longer needed. See Ans. 11,
26-27; Gait, pg. 83 (“The old version is then subject to being garbage
collected.”). Further, as the garbage collection in Gait frees the memory for
writing again in the future, we agree with the Examiner’s finding that
garbage collection constitutes “unmarking” as read-only as recited in claim
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

25

15. See Ans. 26-27 (citing Gait, pg. 83). We therefore sustain the
anticipation rejection of claim 15.

CONCLUSION
We conclude that:
(1) the Examiner incorrectly determined that claim 15 is anticipated
by Quinlan;
(2) the Examiner correctly determined that claim 4 is anticipated by
Rosenblum;
(3) the Examiner correctly determined that claim 4 is anticipated by
Sidebotham;
(4) the Examiner correctly determined that claims 4, 8, 11-14, and 20
are anticipated by Quinlan; and
(5) the Examiner correctly determined that claims 4, 8, 11-15, 20, 23,
and 24 are anticipated by Gait.

DECISION
For the above reasons, the rejection of claim 15 as anticipated by
Quinlan is reversed, and the rejection of claim 4 as anticipated by
Rosenblum, the rejection of claim 4 as anticipated by Sidebotham, the
rejection of claims 4, 8, 11-14, and 20 as anticipated by Quinlan, and the
rejection of claims 4, 8, 11-15, 20, 23, and 24 as anticipated by Gait are
affirmed. At least one ground of rejection for each rejected claim has been
affirmed.
Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

26

Requests for extensions of time in this ex parte reexamination
proceeding are governed by 37 C.F.R. § 1.550(c). See 37 C.F.R. § 41.50(f).

AFFIRMED

Appeal 2012-004252
Reexamination Control Nos. 90/008,903 and 90/009,103
Patent 5,819,292

27

Patent Owner:

Cesari and McKenna, LLP
88 Black Falcon Avenue
Boston, MA 02210

Third Party Requester:

DLA Piper US LLP
2000 University Avenue
E. Palo Alto, CA 94303-2248