Ex Parte Lashley

Patent Trial and Appeal Board

Sep 21, 2015

10941542 (P.T.A.B. Sep. 21, 2015)

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.
10/941,542 09/14/2004 Scott David Lashley BEA920030035US1 9940
46156 7590 09/21/2015
LIEBERMAN & BRANDSDORFER, LLC
802 Still Creek Lane
Gaithersburg, MD 20878-3218
EXAMINER
ALVESTEFFER, STEPHEN D
ART UNIT PAPER NUMBER
2171
MAIL DATE DELIVERY MODE
09/21/2015 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 PATENT TRIAL AND APPEAL BOARD
____________

Ex parte SCOTT DAVID LASHLEY
____________

Appeal 2013-007934
Application 10/941,542
Technology Center 2100
____________

Before JOHN G. NEW, CARL L. SILVERMAN, and CHRISTA P. ZADO,
Administrative Patent Judges.

ZADO, Administrative Patent Judge.

DECISION ON APPEAL
Appellant1 files this appeal under 35 U.S.C. § 134(a) from the
Examiner’s final rejection of claims 1–20.2 We have jurisdiction under 35
U.S.C. § 6(b).
We affirm.

1 Appellant identifies the real party in interest as International Business
Machines Corporation. App. Br. 3.
2 No other claims are pending in the instant Application.
Appeal 2013-007934
Application 10/941,542

2
STATEMENT OF THE CASE3
Claim 1 is illustrative of the subject matter on appeal:
1. A method for crash recovery in a data base management
system (DBMS), the method comprising:
copying a plurality of updated and frequently referenced pages
into a first fast recovery log;
loading a plurality of sequential pages of data as a block in a
single transaction from the first fast recovery log into a bufferpool
responsive to a disruption, said pages formed into blocks, and wherein
said first fast recovery log is a separate area in persistent storage from
a database table;
applying a plurality of logical operations from a logical log to
said pages in said bufferpool to return the DBMS to a transactionally
consistent state; and
allocating a second fast recovery log when said first fast
recovery log is full, including flushing the first fast recovery log to the
database.

THE REJECTION
Claims 1–20 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over Arun (US 5,933,593; issued Aug. 3, 1999), IBM (Efficient Periodic
Database Buffer Flushing Policy to Optimize Recovery and On-Line Write
Input/Output Costs, IBM TECHNICAL DISCLOSURE BULLETIN, vol. 38, issue
no. 11, 281–282 (Nov. 1, 1995)), and Hill (US 6,334,171 B1; issued Dec.
25, 2001). Final Act. 2–30.

3 Rather than repeat the arguments here, we refer to the Appeal Brief and
Reply Brief for the positions of Appellant and the Final Office Action,
Advisory Action, and Answer for the positions of the Examiner.
Appeal 2013-007934
Application 10/941,542

3
ANALYSIS
We have reviewed the evidence and arguments raised by Appellant in
the Appeal Brief and Reply Brief; however, we are not persuaded the
Examiner erred in rejecting claims 1–20.
The Examiner finds Arun teaches a method for crash recovery in a
database management system (DBMS) that is substantially the same as
Appellant’s recited invention. Ans. 7. Specifically, the Examiner finds
Arun teaches the method of claim 1, except for copying frequently
referenced (i.e., hot pages) into a fast recovery log, and allocating a second
fast recovery log when a first fast recovery log is full. Final Act. 2–5.
However, the Examiner finds, and we agree, IBM and Hill make up for the
deficiencies of Arun. Id.
Arun teaches copying pages that have been recently updated into
checkpoint files (i.e., a first fast recovery log) that are stored in persistent
memory, so that these pages may be quickly accessed in the event of a crash.
Id. Arun does not teach also copying frequently referenced pages into the
checkpoint files, as required by claim 1; however, IBM teaches copying hot
pages (i.e., frequently updated pages) to a buffer so that the pages may be
quickly accessed in the event of a crash. Id. at 4–5. The Examiner finds,
and we agree, it would have been obvious to modify Arun in view of IBM to
also copy frequently updated pages into Arun’s checkpoint files. Id.
Arun does not teach “allocating a second fast recovery log when said
first fast recovery log is full, including flushing the first fast recovery log to
the database,” as recited in claim 1. Id. at 5. However, Hill teaches when a
first active storage buffer becomes full, allocating data to a new active
buffer, and flushing the first active buffer. Id. (citing Hill 5:14–24). The
Appeal 2013-007934
Application 10/941,542

4
Examiner finds, and we agree, that it would have been obvious in view of
Hill to allocate a second checkpoint file in Arun when the first checkpoint
file becomes full, and flushing the first checkpoint file. Id.
Appellant argues:
Applicant’s claims are directed to a fast recovery log and
storage of the log itself in persistent storage. Hill does not
teach or suggest this claim limitation. Accordingly, Hill comes
short of teaching a second fast recovery log created in
persistent memory when a first fast recovery log is full.
App. Br. 12; see also id. at 13–15. We are not persuaded by this argument
because the Examiner relies upon Arun, not Hill, for teaching storing a
recovery log (i.e., checkpoint files) in persistent memory. Ans. 6; see also
Arun 3:8–13. “Hill is only relied upon for its teaching that the practice of
allocating a new memory location when a first one becomes full, then
flushing the first memory location to a database, was well known in the art at
the time the invention was made.” Ans. 6.
Appellant argues a skilled artisan would not have been motivated to
modify Arun’s invention to “incorporate the lazy hot page flushing
techniques” in view of IBM because the database buffers in IBM represent
volatile storage, and are not stored in persistent memory. App. Br. 13. This,
too, is not persuasive because the Examiner relies on Arun, rather than IBM,
for storing fast recovery logs in persistent storage. Ans. 4. Although the
Examiner relies on IBM for its teaching to store frequently updated (i.e., hot)
pages, the Examiner relies on Arun for teaching that the storage is in
persistent memory. Final Act. 4.
Appellant also argues the Examiner does not address the claim
limitations pertaining to the term “block” in claims 1, 12, and 19:
Appeal 2013-007934
Application 10/941,542

5
None of Fig. 6, 2:64–3:7 or 7:61–8:15 address the claim term
“block” or “blocks” and how they support the claims page or
pages. In fact, a thorough review shows that Arun does not
employ this claim limitation in their Specification.
App. Br. 15. We disagree with Appellant, and find the Examiner has met
the burden of showing Arun teaches the claim limitations pertaining to the
term “block.” During examination of a patent application, pending claims
are given their broadest reasonable construction consistent with the
Specification. In re Am. Acad. of Sci. Tech Ctr., 367 F.3d 1359, 1364 (Fed.
Cir. 2004); In re Prater, 415 F.2d 1393, 1404-05 (CCPA 1969). A claim
construction “that excludes a preferred embodiment from the scope of the
claim is rarely, if ever, correct.” Accent Packaging, Inc. v. Leggett & Platt,
Inc., 707 F.3d 1318, 1326 (Fed. Cir. 2013) (quotations and citation omitted).
Appellant’s Specification provides that fast recovery log pages are stored
sequentially as a single block to increase recovery time by avoiding multiple
random I/Os of separate pages:
The approach proposed for this, described in more detail below,
is to provide and employ a separate area in persistent storage to
contain copies of the pages 24 needed for crash recovery. This
storage area is termed a “fast recovery log” and, unlike the
contents of the logical log 18, the pages in the fast recovery log
are stored such that they can be loaded into the buffer pool 16
sequentially. This provides improved crash recovery by
replacing multiple relatively slow random I/Os with as few as a
single large block, sequential I/Os.
Spec. 11:6–12. Accordingly, we conclude the term “block” encompasses
pages of data that are stored sequentially. We therefore agree with the
Examiner’s finding that Arun’s teaching of writing modified records to
checkpoint files “in one continuous, uninterrupted bit stream,” which Arun
Appeal 2013-007934
Application 10/941,542

6
teaches is much faster than scattered I/O operations (as does Appellant’s
Specification), is properly interpreted as teaching “blocks” of memory.
Ans. 8 (citing Arun 5:31–49).
For the foregoing reasons, we are not persuaded the Examiner erred in
rejecting claims 1–5 and 12–19, which Appellant argues as a group
(hereinafter, “Group I”). Furthermore, we are not persuaded the Examiner
erred in rejecting claims 6–11 and 20, which Appellant argues as a group,
and for which Appellant does not raise separate patentability arguments
from Group I. See generally App. Br. 16–21.

DECISION
The Examiner’s decision rejecting claims 1–20 is affirmed.
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)(iv).

AFFIRMED

ELD