Ex Parte Yim et al

Patent Trial and Appeal Board

Feb 11, 2013

11457633 (P.T.A.B. Feb. 11, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte WAI C. YIM and CHARLES BINFORD
____________

Appeal 2010-004964
Application 11/457,633
Technology Center 2100
____________

Before BRUCE R. WINSOR, STANLEY M. WEINBERG,
and DANIEL N. FISHMAN, Administrative Patent Judges.

WINSOR, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from a final rejection of
claims 1-20, which constitute all the claims pending in this application. We
have jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.

Appeal 2010-004964
Application 11/457,633

2
STATEMENT OF THE CASE
[Appellants’] invention relates . . . to data storage and
back-up solutions that allow recovery of replicated data, and,
more particularly, . . . [to] write order control in a data storage
system that implements asynchronous replication to provide
data protection for a host or primary site with data targets
located at a remote or secondary site.
Spec. ¶ [0001]. Claims 1, 8, and 14, which are illustrative of the invention,
read as follows:
1. A method for ordering writes in a remote data
storage system such that the ordering of the writes is preserved
in the remote data storage system, the method comprising:
receiving a plurality of writes to a volume in a first data
storage resource;
operating a write manager to group each of the writes
into one of a plurality of concurrency groups with a replica
write for each of the writes including a concurrency group
identifier identifying the one of the concurrency groups to
which the write was assigned, wherein the writes assigned to
one of the plurality of concurrency groups are issued together
before any are completed;
transferring the replica writes over a communication
network to the remote data storage system;
processing the concurrency group identifiers of the
transferred replica writes with a write order manager; and
with the write order manager, issuing a subset of the
replica writes concurrently to a target volume on a second data
storage resource in the remote data storage system based on the
concurrency group identifiers.
8. A data storage system implementing asynchronous
replication with write order control for data backups,
comprising:
a primary site comprising a server running an application
that issues a plurality of writes, a storage controller running a
write manager, a primary data storage resource, and an
Appeal 2010-004964
Application 11/457,633

3
asynchronous log, wherein the write manager stamps each of
the writes with a concurrency group identifier prior to storing
the writes in the asynchronous log, and wherein writes stamped
with the same concurrency group identifier are issued together
by the application before any are completed; and
a secondary site linked to the primary site comprising
memory for storing the writes along with the concurrency group
identifiers, a secondary data storage resource with target
volumes corresponding to application volumes on the local data
storage resource, and a storage controller running a write order
manager, wherein the write order manager concurrently issues a
set of the writes stored in the memory based on the concurrency
group identifiers.
14. A computer based method for controlling write
order in a data storage system implementing asynchronous
replication, comprising:
providing an asynchronous log in memory at a primary
computer system;
issuing host writes to an application volume in data
storage local to the primary computer system with an
application running on the primary computer system;
tagging each of the issued host writes with a concurrency
group identifier defining a concurrency group for the issued
host writes; and
queuing the tagged writes in the asynchronous log for
asynchronous replication at a secondary computer system
linked to the primary computer system.

Claims 1-7 stand rejected under 35 U.S.C. § 102(b) as being
anticipated by Crowley (US 2005/0050286 A1; issued Mar. 3, 2005). Ans.
3-5.
Claims 8-20 stand rejected under 35 U.S.C. § 102(b) as being
anticipated by Burton (US 2005/0091391 A1; issued Apr. 28, 2005). Ans.
5-9.
Appeal 2010-004964
Application 11/457,633

4
Rather than repeat the arguments here, we refer to the Brief (Br. filed
Aug. 10, 2009)1 and the Answer (Ans. mailed Nov. 17, 2009) for the
respective positions of Appellants and the Examiner. Only those arguments
actually made by Appellants have been considered in this decision.
Arguments that Appellants did not make in the Brief have not been
considered and are deemed to be waived. See 37 C.F.R. § 41.37(c)(1)(vii)
(2010).

CLAIMS 1-7
Issue
The Examiner finds that Crowley discloses all of the limitations
recited in claim 1. Ans. 3-4 (citing Crowley, Figs. 4 (refs. 430, 440, 450), 5
(ref. 540); ¶¶ [0023], [0031] (ll. 8-13)). Appellants contend, inter alia, that
Crowley does not disclose a “concurrency group” as recited in claim 1. Br.
6-7. The dispositive issue presented by Appellants’ contentions is as
follows:2 Does Crowley disclose “operating a write manager to group each
of the writes into one of a plurality of concurrency groups with a replica
write for each of the writes including a concurrency group identifier
identifying the one of the concurrency groups to which the write was
assigned” (emphases added) (herinafter the “operating step”), as recited in
claim 1?

1 We have not considered the Brief filed June 3, 2009, as it is deemed to
have been superseded by the Brief filed August 10, 2009.
2 Appellants’ contentions regarding claims 1-7 present additional issues.
Because we are persuaded of error with regard to the identified issue, which
is dispositive as to claims 1-7, we do not reach the additional issues.
Appeal 2010-004964
Application 11/457,633

5
Analysis
The Examiner finds that Crowley discloses assigning a sequence
number to data to be written to storage and that assigning the sequence
number reads on the operating step. Ans. 3 (citing Crowley, ¶ [0023]; Fig. 4
(ref. 430)). Appellants contend that in Crowley the sequence number is not
a concurrency group identifier because a sequence number is assigned to an
individual write rather than to a group of writes. Br. 6-7. Appellants further
contend the ordinary and customary meaning of the term “group” in claim 1
is that a group has more than one member. Br. 6 (citing MERRIAM-WEBSTER
ONLINE DICTIONARY. Retrieved June 1, 2009, from http://www.merriam-
webster.com/dictionary/ group).
The Examiner explains that
the broadest reasonable interpretation of the word group is ‘any
number of entities (members) considered as a unit’ (WordNet-
Online). . . . [A] group can contain any number of elements including
zero. Although Appellant may intend for each of the concurrency
groups to contain more than one member, this is not explicitly
reflected in the claim language.

Ans. 10.

Although we agree with the Examiner that “group,” standing alone,
may encompass groups with one or no members, the terms of the claim are
to be construed in light of Appellants’ Specification, see In re Morris, 127
F.3d 1048, 1054 (Fed. Cir. 1997), and the context of the term in the claim
itself, see Hockerson-Halberstadt, Inc. v. Converse Inc., 183 F.3d 1369,
1374 (Fed. Cir. 1999) (“[p]roper claim construction ... demands
interpretation of the entire claim in context, not a single element in
isolation.”). Furthermore, in construing the claims we must consider all of
Appeal 2010-004964
Application 11/457,633

6
the words of the claims. In re Wilson, 424 F.2d 1382, 1385 (CCPA 1970).
In construing the term “group” we note that “group” is used in claim 1
both as a verb and as a noun. Further, we note that the noun “group” is
modified by “concurrency.” Pertinent dictionary definitions of “group” are
as follows:
group . . . n . . . 1: two or more figures forming a complete unit
in a composition 2 a: a number of individuals assembled
together or having some unifying relationship b: an assemblage
of objects regarded as a unit . . .
group vt . . . 1: to combine in a group 2: to assign to a group:
classify . . . .
MERRIAM-WEBSTER’S COLLEGIATE DICTIONARY 515 (10th ed. 1999); accord
Br. 6. The pertinent dictionary meaning of concurrency is “concurrence
. . . n . . . 1 a: the simultaneous occurrence of events or circumstances . . .
concurrency . . . n . . . : concurrence . . . .” Id. at 240; see also MICROSOFT
COMPUTER DICTIONARY 122 (5th ed. 2002) (“concurrent adj. Of,
pertaining to or characteristic of a computer operation in which two or more
processes (programs) have access to the microprocessor’s time and are
therefore carried out nearly simultaneously. . . .”).
In the context of claim 1, the “group” recited is a “concurrency group”
formed by “group[ing] each of the writes into one of a plurality of [the]
concurrency groups.” One of ordinary skill in the art would understand a
“concurrency group” to be a group of events or circumstances that occur, at
least in part, simultaneously or nearly so. This is consistent with the
Appellants’ Specification which states that “[c]oncurrency is generally
defined by writes that are issued together by the host or host application
before any are completed.” Spec. ¶ [0017] (ll. 5-6). For events or
circumstances (note the plural in the definition) in a “concurrency group” to
Appeal 2010-004964
Application 11/457,633

7
occur simultaneously there must be at least two, otherwise, the word
“concurrency” has no meaning in the claim. See Wilson, 424 F.2d at 1385.
Indeed, the Examiner relatedly states that “[t]he broadest reasonable
interpretation of ‘concurrently’ is that the writes overlap in duration.” Ans.
13. For writes in a “concurrency group” to “overlap in duration,” or be
“issued together,” there must be at least two writes.
We conclude that, in the context of claim 1, a “concurrency group” is
a group of two or more writes that occur, at least in part, simultaneously or
together, or nearly so, before either write is complete. We agree with
Appellants that Crowley’s “sequence number” does not read on a
“concurrency group” because it only identifies a single write. See Br. 6-7.
Therefore, the Examiner has failed to demonstrate that Crowley discloses the
operating step.
Summary
We find error in the rejection of claim 1, as it is based on an incorrect
claim construction. Accordingly, we will not sustain the rejection of claim
1, and claims 2-7, which depend from claim 1 and were rejected based on
the same incorrect claim construction.

CLAIMS 8-13
Issue
The Examiner finds that Burton discloses all of the limitations recited
in claim 8. Ans. 5-6 (citing Burton, Figs. 1 (ref. 18), 2, 3 (data set 0); ¶¶
[0038], [0048], [0060]). Appellants contend, inter alia, that “Burton does
not disclose ‘wherein the write order manager concurrently issues a set of
the writes stored in the memory based on the concurrency group
Appeal 2010-004964
Application 11/457,633

8
identifiers.’” Br. 10. The dispositive issue presented by Appellants’
contention is as follows:3 Does Burton disclose that “the write order
manager [at the secondary site] concurrently issues a set of the writes stored
in the memory based on the concurrency group identifiers” (emphasis
added) (herinafter the “concurrent write limitation”), as recited in claim 8?
Analysis
The Examiner relies on Burton at ¶ [0038], lines 18-21, as disclosing
the concurrent write limitation. Ans. 6, 13. Burton discloses that “at step 6,
[(see Fig. 2)] when the secondary host 1 has received all of the blocks of a
data set, the secondary host 1 writes the blocks of the data set to the
secondary source VLUN.” Burton, ¶ [0038] (ll. 18-21). Appellants contend
that the passage of Burton relied upon by the Examiner “merely states that
the secondary host writes the blocks of a data set to the secondary storage
after it has received all of the blocks of the data set from the primary host,
and does not disclose writing the blocks of a data set concurrently.” Br. 10.
We agree with Appellants for the reasons stated by Appellants. Id.
Summary
We find that the Examiner has not articulated a prima facie case that
Burton discloses the concurrent write limitation. Therefore, we find error in
the rejection of claim 8 as anticipated by Burton. Accordingly, we will not
sustain the rejection of claim 8 and claims 9-13, which depend from claim 8.

3 Appellants’ contentions regarding claims 8-13 present additional issues.
Because we are persuaded of error with regard to the identified issue, which
is dispositive as to claims 8-13, we do not reach the additional issues.
Appeal 2010-004964
Application 11/457,633

9
CLAIMS 14-20
Issue
The Examiner finds that Burton discloses all of the limitations recited
in claim 14. Ans. 7-8 (citing Burton, Figs. 2, 12; ¶¶ [0038] (ll. 18-21),
[0040], [0048]). Referring to arguments made regarding claim 8, Appellants
contend, inter alia, that “Burton does not disclose tagging each of the issued
host writes with a concurrency group identifier defining a concurrency group
for the issued host writes” (hereinafter the tagging step). Br. 11.
Appellants further contend that because Burton does not disclose the tagging
step, Burton does not disclose “queuing the tagged writes (i.e., the writes
tagged with concurrency group identifiers) in the asynchronous log for
asynchronous replication at a secondary computer system linked to the
primary computer system” (hereinafter “the queuing step”). Id. The issue
presented by Appellants’ contentions is as follows: Does Burton disclose
the tagging step recited in claim 14?
Analysis
The Examiner relies on Burton at ¶ [0048] as disclosing the tagging
step. Ans. 8; see also Ans. 13-14. Referring to a recitation in claim 8,
which is similar to the tagging step recited in claim 14, Appellants argue
that:
[a]lthough Burton does disclose that writes sent to a secondary
host between sync commands are stamped with a common data
set ID, Burton does not provide any indication that a data set ID
is equivalent to a concurrency group identifier. Conversely, the
concurrency group identifier disclosed in claim 8 is specifically
used to identify writes that are issued together (by the
application running on the server at the primary site) before any
are completed, such that the writes of a concurrency group may
be issued concurrently (i.e., in parallel) at a secondary site,
Appeal 2010-004964
Application 11/457,633

10
thereby increasing the bandwidth of the secondary data storage
resource.
Br. 9.
The Examiner responds:
[T]he claim only requires that writes issued together be
assigned an identifier and that these writes be issued
concurrently based on the identifier. . . . [Burton’s] data set ID
identifies writes belonging to the same group (¶ 0040, lines 1-
3) and the writes belonging to the same data set are issued
together by the application before they are completed at the
secondary host (figure 2; ¶ 0038, lines 18-21). These two
teachings show that writes belonging to the same group are
given the same data set ID and these writes are completed at the
secondary device once the secondary device has received all the
pending writes. These writes are issued concurrently, because
they are all concurrently pending at the secondary site until they
are all received.
Ans. 13-14. We agree with the Examiner’s claim construction and mapping
of the claim onto Burton’s disclosure.
Appellants further argue, again with reference to a similar recitation in
claim 8, that a concurrency group, as recited, is a group of writes “issued
together by the application before they are completed at the primary site.”
Br. 10. However, as pointed out by the Examiner, claim 8 (and 14) do not
recite that a concurrency group is a group of writes “issued together by the
application before they are completed at the primary site.” Ans. 12-13.
We find Appellants arguments regarding the tagging step do not
persuasively rebut the Examiner’s findings and explanations. As
Appellants’ arguments regarding the queuing step merely rely on the
arguments for the tagging step, they are similarly unpersuasive. We find
that the Examiner’s reading of claim 14 on Burton’s disclosure to be
Appeal 2010-004964
Application 11/457,633

11
reasonable and not persuasively rebutted by the Appellants. Accordingly,
we agree with the Examiner for the reasons stated by the Examiner.
Summary
We find that Appellants have not demonstrated error in the rejection
of claim 14. Accordingly, we will sustain the rejection of claim 14, and
claims 15-20, which depend from claim 14 and were not separately argued
with particularity. Br. 11.

ORDER
The decision of the Examiner to reject claims 14-20 is affirmed.
The decision of the Examiner to reject claims 1-13 is reversed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)). See 37 C.F.R.
§ 1.136(a)(1)(iv) (2010).

AFFIRMED-IN-PART

rwk