James M. Higgins et al.

Patent Trials and Appeals Board

Sep 25, 2019

13862314 - (D) (P.T.A.B. Sep. 25, 2019)

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.
13/862,314 04/12/2013 James M. Higgins 058752-01-5201-US 5737
82474 7590 09/25/2019
Morgan, Lewis & Bockius LLP (PH)(SanDisk)
1701 Market Street
Philadelphia, PA 19103-2921
EXAMINER
SAIN, GAUTAM
ART UNIT PAPER NUMBER
2135
NOTIFICATION DATE DELIVERY MODE
09/25/2019 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
judith.troilo@morganlewis.com
phpatentcorrespondence@morganlewis.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
_____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
_____________

Ex parte JAMES M. HIGGINS, JAMES M. KRESSE, RYAN JONES, and
MARK DANCHO1
_____________

Appeal 2017-010643
Application 13/862,314
Technology Center 2100
______________

Before ROBERT E. NAPPI, JOHNNY A. KUMAR, and CARL L.
SILVERMAN Administrative Patent Judges.

NAPPI, Administrative Patent Judge.

DECISION ON APPEAL
This is a decision on appeal under 35 U.S.C. § 134(a) of the Final
Rejection of claims 1 through 5, 7, 10 through 15, 17, 19, and 20.
We reverse.
INVENTION
The invention is directed to a method of operation of a storage control
system, including: receiving a recycle write from a recycle write queue;
receiving a host write from a host write queue; and scheduling the recycle
write and the host write for writing to a memory device. The writes
scheduler maintains a target ratio of recycle writes and host writes and

1 According to Appellant, the real party in interest is Sandisk Technologies
LLC. App. Br. 4.
Appeal 2017-010643
Application 13/862,314

2

adjusts the target ratio based upon the pool of erased memory that can be
written to. Spec., paras. 7, 62, 72.
Claim 1 is illustrative of the invention and is reproduced below:
1. A method of operation of a storage control system for
a memory device comprising:
receiving at least one recycle write from a recycle
write queue;
receiving at least one host write from a host write
queue;
identifying a target recycle ratio in accordance
with a current size of a spare pool of erased non-volatile
memory blocks available to be written to, wherein the
target recycle ratio is a targeted ratio of recycling page
writes to host page writes;
placing the at least one recycle write and the at
least one host write in a write operation queue in
accordance with the target recycle ratio;
dispatching the at least one recycle write and the at
least one host write from the write operation queue for
writing data to the memory device;
during operation of the storage control system,
updating the target recycle ratio in accordance with a
change in the current size of the spare pool; and
after updating the target recycle ratio:
placing one or more recycle writes and one
or more host writes in the write operation queue in
accordance with the updated target recycle ratio;
and
dispatching the one or more recycle writes
and the one or more host writes from the write
operation queue for writing data to the memory
device.

REJECTIONS AT ISSUE
The Examiner rejected claims 1 through 5, 7, 10 through 15, 17, and
20 under 35 U.S.C. § 103 as being unpatentable over Ebsen (US
Appeal 2017-010643
Application 13/862,314

3

2012/0278530 A1, published Nov. 1, 2012), Kang (US 8,725,931 B1, issued
May 13, 2014) and Post (US 2011/0055455 A1, published Mar. 3, 2011).
Final Act 2–14.2
The Examiner rejected claim 19 under 35 U.S.C. § 103 as being
unpatentable over Ebsen, Kang, Post and Higashino (US 2009/0179707 A1,
published Jul. 16, 2009). Final Act 14–15.
ANALYSIS
Appellant argues that the Examiner’s rejection of independent claims
1 and 11 is in error as the combination of the references does not teach
identifying a target recycle ratio in accordance with a current size of a spare
pool of erased non-volatile memory blocks to be written to as recited in each
of the independent claims. App. Br. 12–18, Reply Br. 7–10. Appellant
argues that Ebsen teaches a priority ratio, but never discusses this ratio as the
ratio of recycling writes to host page writes as claimed. App. Br. 13–14.
Further, Appellant acknowledges that Kang teaches a throttling ratio based
upon the number of pages of invalid data in a memory block undergoing
garbage collection but does not teach that the ratio is adjusted according to
the current size of a spare pool of erased non-volatile memory blocks as
claimed. App. Br. 14 (citing Kang, col. 2). Finally, Appellant argues that
Post teaches adjusting the recycle ratio in accordance with a block’s
proportion of valid data, not the current size of a spare pool of erased non-
volatile memory blocks as claimed. App. Br. 15 (citing, paras. 43, 60. 69,
70, 71. Concerning Post’s teaching in Figure 3, Appellant states:

2 Throughout this Opinion, we refer to the Appeal Brief, filed February 17,
2017 (“App. Br.”), the Reply Brief, filed August 14, 2017 (“Reply Br.”), the
Examiner’s Answer, mailed June 16, 2017 (“Answer”), and the Final Office
Action, mailed July 20, 2016 (“Final Act”).
Appeal 2017-010643
Application 13/862,314

4

In Post, the number of memory locations in an erased
state is used to determine which of three modes of garbage
collection, incremental, full-on, or no garbage collection,
should be used (see Figure 3, reproduced below). In particular,
the number of memory locations in an erased state is compared
to a predetermined threshold (i.e., threshold 302), and the result
of that comparison is used to enable or disable incremental
garbage collection. However, Post does not teach or suggest,
“identifying a target recycle ratio in accordance with a current
size of a spare pool of erased non-volatile memory blocks
available to be written to.” Post’s threshold, based on memory
locations in the die, is merely used to determine if incremental
garbage collection should be initiated.
App. Br. 16
The Examiner finds Kang teaches a throttling ratio for garbage
collection, which the Examiner equates to the claimed recycling ratio and
finds that Kang teaches the ratio can be adjusted based upon monitored
conditions including wear level, erase counts. Answer 17 (citing Kang, col.
4, ll. 1–15, 22–27, col. 2, ll. 26–37, col. 3, ll. 24–31, and Figs. 3, 4A, 4B);
Final Act. 4. Further, the Examiner finds that
Post discloses that number of memory locations in a die of
NVM that are in erased state, para 0039, Fig. 3. Fig. 4 shows
two blocks in an erased state 410, 412, to program additional
data into blocks 410, para 0042, which is interpreted as claimed
spare pool since the blocks in erase state are available to be
written to and are spare storage since they are not being
currently used to store data.
Answer 18. Based upon these findings, the Examiner concludes it would be
obvious to adjust the throttling ratio based upon the number of pages with
invalid data undergoing garbage collection. Answer 18, Final Act. 5.
We disagree with the Examiner’s conclusion. Each of the
independent claims, 1 and 11, recites “identifying a target recycle ratio in
Appeal 2017-010643
Application 13/862,314

5

accordance with a current size of a spare pool of erased non-volatile memory
blocks available to be written to, wherein the target recycle ratio is a targeted
ratio of recycling page writes to host page writes.” We concur with the
Examiner that Kang teaches adjusting a recycle ratio and that Post teaches a
number of erased non-volatile memory blocks available to be written.
However, we do not find that either of the references teaches or suggests the
ratio is adjusted in accordance with the size of the spare pool. As argued by
the Appellant, Post uses the number of erased state blocks to determine
which of three modes to operate in. See Post, para. 69 and Fig. 3. In Post’s
no garbage collection mode and the full-on garbage collection mode there is
no discussion of a ratio, as there is either no recycle writes or no host writes.
(see Post Fig.3) In the incremental garbage collection mode, the ratio of
garbage collection is used and adjusted based upon the proportion of valid
data. See Post, para. 70. The Examiner has not adequately explained, nor is
it apparent to us, how this aspect of Post teaches or suggests adjusting the
recycle ratio as claimed. Accordingly, we do not sustain the Examiner’s
rejection of independent claims 1 and 11 or dependent claims 2 through 5, 7,
10, 12 through 15, 17, and 20.
The Examiner has not shown that the teachings of Higashino, used in
the rejection of claim 19, make up for the deficiencies, noted in the rejection
of independent claims 1 and 11. Accordingly, we similarly do not sustain
the Examiner’s rejection of claim 19.

Appeal 2017-010643
Application 13/862,314

6

DECISION
The decision of the Examiner to reject claims 1 through 5, 7, 10
through 15, 17, 19, and 20 is reversed.

Claims
Rejected
Basis Affirmed Reversed
1–5, 7, 10–15, 17
and 20
§ 103 Ebsen, Kang,
and Post
1–5, 7, 10–15, 17, and 20
19 § 103 Ebsen, Kang,
Post, and Higashino
19
Outcome 1–5, 7, 10–15, 17, 19, and 20

REVERSED