Ex Parte Loh et al

Patent Trial and Appeal Board

Oct 31, 2016

13567945 (P.T.A.B. Oct. 31, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
13/567,945 08/06/2012 Gabriel H. Loh
109712 7590 11/02/2016
Advanced Micro Devices, Inc.
c/o Davidson Sheehan LLP
8834 North Capital of TX Hwy
Suite 100
Austin, TX 78759
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
1458-110209 5029
EXAMINER
RIZK, SAMIR W ADIE
ART UNIT PAPER NUMBER
2112
NOTIFICATION DATE DELIVERY MODE
11/02/2016 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):
docketing@ds-patent.com
beatrice. zepeda@ds-patent.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte GABRIEL H. LOH, JAMES M. O'CONNOR,
BRADFORD M. BECKMANN, and MICHAEL IGNATOWSKI
Appeal2015-007443
Application 13/567,945
Technology Center 2100
Before JUSTIN BUSCH, SCOTT E. BAIN, and ALEX S. YAP,
Administrative Patent Judges.
YAP, Administrative Patent Judge.
DECISION ON APPEAL
Appellants 1 appeal under 35 U.S.C. § 134(a) from the Examiner's
final rejection of claims 1--40, which are all the claims pending in this
application. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM-IN-PART.
1 According to Appellants, the real party in interest is Advanced Micro
Devices, Inc. (App. Br. 1.)
Appeal2015-007443
Application 13/567,945
STATEMENT OF THE CASE
Introduction
Appellants' invention "relates to memory devices, and more
particularly, to stacked memory devices." (Aug. 6, 2012 Specification
("Spec.") i-f 2.) Claim 1 is representative and is reproduced below:
1. An integrated circuit (IC) package comprising:
memory cell circuitry; and
a set of one or more logic layers electrically coupled to
the memory cell circuitry, the set of one or more logic layers
comprising a metadata manager and a memory interface, the
memory interface coupled to the metadata manager and
coupleable to a device external to the IC package, and the
metadata manager to manage metadata stored at the IC package
and which is associated with operational data stored at the
memory cell circuitry for the device.
Prior Art and Rejections on Appeal
The follov,ring table lists the prior art relied upon by the Examiner in
rejecting the claims on appeal:
Blankenship
Call et al.
("Call")
Strasser et al.
("Strasser")
US 2010/0070696 Al Mar. 18, 2010
US 8,700,951 Bl Apr. 15, 2014
US 2014/0108891 Al Apr. 17, 2014
Claims 1-10 and 12--40 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over Call in view of Blankenship. (See Final Office
Action (mailed Oct. 31, 2014) ("Final Act.") 5-14.)
2
Appeal2015-007443
Application 13/567,945
Claim 11 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Call in view of Blankenship, and further in view of
Strasser. (See Final Act. 14--15.)
ANALYSIS
We have reviewed the Examiner's rejections in light of Appellants'
arguments that the Examiner has erred. We are not persuaded that the
Examiner erred in rejecting claims 1-3, 6, 7, and 9--40. We are persuaded
that the Examiner erred in rejecting claims 4, 5, and 8.
Claims 1-3, 6, 7, and 9--40
With respect to claim 1, the Examiner finds that Blankenship teaches
or suggests the limitations of claim 1 but does not teach a metadata manager
to manage metadata stored at the integrated circuit (IC) package. (Final Act.
5---6.) The Examiner, however, finds that Call teaches or suggests a metadata
manager and that:
It would have been obvious to one of ordinary skill in the art at
the time the invention was made to modify the teaching of Call
[and] the teaching of Blankenship that comprises IC package
comprising memory array (cells) and memory interface
controller.
This modification would have been obvious to one of ordinary
skill in the art, at the time the invention was made, because one
of ordinary skill in the art would have recognized the need for
low power and compact electronics and increased data
reliability.
(Final Act. 6; Ans. 4.) Appellants disagree and contend that the combination
"would not result in the particular combinations of features recited in
claim 1" because "Blankenship describes a memory system ... whereas Call
3
Appeal2015-007443
Application 13/567,945
describes a disk storage system [and] if one of ordinary skill in the art were
to combine the teachings of Call and Blankenship, the end result would be a
computing system employing a memory system as taught by Blankenship
and a separate disk storage system as taught by Call." (App. Br. 5.)
Appellants further contend that:
there is no evidence of any motivation for one of ordinary skill
in the art to make such modifications, if such modifications
could even be implemented at all.
[Moreover,] Blankenship already provides for an on-chip ECC
approach that provides low power consumption, compact
electronics, and increased data reliability, and it has not been
demonstrated that the incorporation of any of the teachings of
Call would improve upon what is already achieved by
Blankenship. Conversely, there is no suggestion in Call or any
evidence showing that it was well known at the time of
invention that the incorporation of the teachings of Call into an
IC package such as the one taught by Blankenship (if such
combination is even achievable by one of ordinary skill) would
result in lower power consumption, more compact electronics,
or increased data reliability.
(App. Br. 5---6, emphasis added; Reply 1-2.) The Examiner disagrees and
further states that Blankenship also supports the Examiner's rationale for the
combination. (Ans. 4--5, citing to Blankenship i-f 4 ("As portable electronic
devices become smaller, less expensive, and more power efficient, however,
there has been an increased demand for low power and compact electronic
systems in which DRAM is commonly used. Consequently, systems and
methods are needed to reduce the size, power and cost of memory
systems.").)
Appellants have not persuaded us that the Examiner erred. We find
that the Examiner provides sufficient articulated reasoning having a rational
4
Appeal2015-007443
Application 13/567,945
underpinning, such that a person of ordinary skill in the art would have been
motivated to combine the teachings of Blankenship and Call. (Id.) See KSR
Int'! Co., v. Teleflex, Inc., 550 U.S. 398, 415, 418 (2007). Appellants'
contention that one of ordinary skill in the art would not be able to combine
the teachings of Blankenship and Call because "Blankenship describes a
memory system ... whereas Call describes a disk storage system" is mere
attorney argument and a conclusory statement that is unsupported by factual
evidence, and, thus is entitled to little probative value. In re Geisler, 116
F.3d 1465, 1470 (Fed. Cir. 1997); In re De Blauwe, 736 F.2d 699, 705 (Fed.
Cir. 1984). Moreover, the Examiner does not propose physically combining
Call with Blankenship, but merely Call' s teaching of a metadata manager
with Blankenship's teaching of stacked memory layers on an integrated
circuit. In re Keller, 642 F.2d 413, 425 (CCPA 1981) ("The test for
obviousness is not whether the features of a secondary reference may be
bodily incorporated into the structure of the primary reference . . . . Rather,
the test is what the combined teachings of the references would have
suggested to those of ordinary skill in the art."); see also In re Sneed, 710
F.2d 1544, 1550 Fed. Cir. 1983 (citing Orthopedic Equip. Co. v. United
States, 702 F.2d 1005, 1013 (Fed. Cir. 1983); In re Andersen, 391F.2d953,
958 (CCPA 1968)) ("[I]t is not necessary that the inventions of the
references be physically combinable to render obvious the invention under
review."). Furthermore, we agree with the Examiner's finding that one of
ordinary skill in the art would be able to combine teachings in Call
(regarding the management of metadata) with the teachings of Blankenship
(regarding an "IC package comprising memory array (cells) and memory
interface controller"). (Final Act. 5---6; Ans. 4--5.) Specifically, we agree
5
Appeal2015-007443
Application 13/567,945
with the Examiner that a person of ordinary sill in the art would be able to
implement the metadata management of Call in the Memory Controller and
Interface IC 302 of Blankenship. (Final Act. 5---6.) Figure 3 of Blankenship
is reproduced below.
'\ x~
::;~~ ., j)~1
"""') ~· f .,.,~
~ ....................................... ': .':, :;>"""""""""".,:.""""""""""""""""""""""""""<:. l t:-; ..
~:
~~~~'«')
;>.,~~ .. .:~'
- ' ,............................................................... • ......................................................................................................................... \. ... )o.. , :f ....... ,~ :-,_;.;.::.::.:-...~ ... ~ , ~ ~ ~ ~ ~
~~ ~~,~;.:;::., ~J L J
:~~~ .·.··1~-~r ;~~~~~:-~~~:~~~-:>::~~~:~~-----------------w ------------------w ~ ~ ~l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~m I
L:::::::::::::::::::::::.::.::::::::::::::::::::::::r:::::::::::::::::::::.J
s:~~} ......
Fig.3
Figure 3 depicts "a schematic of a chipset for a packaged DRAM memory
system according to a preferred embodiment." (Blankenship i-f 10.) We
agree with the Examiner that Call teaches or suggests that its metadata
management can be applied to both "mechanical hard disk and solid state
disk (SSD) i.e.[,] integrated circuit (IC) FLASH-based disk." (Call,
Abstract.) For example, Call also teaches metadata management of IC
memones:
As shown [in FIG. 1], a storage subsystem 140 includes a
controller 150 and one or more non-volatile solid-state memory
6
Appeal2015-007443
Application 13/567,945
arrays 160. The arrays 160 may comprise a plurality of solid-
state storage devices coupled to the controller 150. The arrays
may comprise, for example, flash integrated circuits[.] Other
forms of storage (e.g., battery backed-up volatile DRAM or
SRAM devices, magnetic disk drives, etc.) may additionally or
alternatively be used.
(Call, 2: 16-3 3, emphases added.) Therefore, contrary to Appellants'
contention that one of ordinary skill in the art would not be able to combine
the teachings of Blankenship and Call because "Blankenship describes a
memory system ... whereas Call describes a disk storage system" (App. Br.
5), we agree with the Examiner that one of ordinary skill in the art would be
able to implement Call's metadata manager in Blankenship. (Ans. 5.)
We also do not agree with Appellants' contention that "Blankenship
already provides for an on-chip ECC approach ... , and [therefore,] it has
not been demonstrated that the incorporation of any of the teachings of Call
would improve upon what is already achieved by Blankenship." (App. Br. 5,
emphasis added.) Blankenship; however; merely notes in passing that "[ o ]ne
could also place Error Correction Circuitry (ECC) along with the
redundancy for repair block 350." (Blankenship i-f 35.) It does not describe
how ECC is to be implemented and how to improve on it. Call, on the other
hand, teaches that the "spare area[, which is used to store metadata,] may
include metadata such as that used for ECC purposes" and how to handle
uncorrectable ECC error. (Call, 5:9-10; see also id. at 3:40-45, 4:7-13, 30-
37.)
Claim 7 recites that "the metadata manager is to perform at least one
metadata management operation in response to a metadata command from
the device." As discussed above, the Examiner finds that the metadata
manager of Call can be implemented in the Memory Controller and Interface
7
Appeal2015-007443
Application 13/567,945
IC 302 of Blankenship. (Final Act. 6; Blankenship, FIG. 3.) According to
the Examiner, the metadata manager is also "coupleable to a device external
to the IC package, (Figure [3], (304) in Blankenship)" and the metadata
manager would "perform at least one metadata management operation in
response to a metadata command from the device" external to the IC
package (304) as shown in Figure 3. (Final Act. 9.) Appellants contend that
"the 'metadata manager' of Call is paradoxically both part of the IC package
(that is, internal to the IC package) and external to the IC package." (App.
Br. 8-9; Reply 5.) Appellants have not persuaded us that the Examiner
erred. As noted by the Examiner, the metadata manager of Call would be
implemented in the Memory Controller and Interface IC 302 (of
Blankenship), which is in the IC package and Appellants have not persuaded
us that the metadata manager is "external to the IC package."
Claim 10 recites that "the metadata comprises memory log
information representing accessed memory locations; and the at least one
metadata management operation comprises at least one of: configuring at
least one parameter of a memory logging operation to be performed by the
metadata manager; and accessing the memory log information for output to
the device." Appellants contend that neither Figures 4A nor 4B, cited in the
Final Action, "teaches that at least one parameter of this operation is
configured, much less as part of a metadata management operation, as
provided by claims 1 and 10." (App. Br. 10.) The Examiner clarifies that
Call teaches a "controller (Figure 1, ref. (150)[)] to provide for metadata
management operation for configuration of parameter of a memory logging
operation." (Ans. 7.) Appellants do not reply. Appellants have not
8
Appeal2015-007443
Application 13/567,945
persuaded us that the Examiner erred and we agree with the Examiner's
finding regarding this claim.
Claim 11 recites that "the metadata comprises garbage collection
attributes; and the at least one metadata management operation comprises
modifying at least one of the garbage collection attributes." The Examiner
finds that while "Call [and] Blankenship teach substantially all the
limitations in claims 1 and 7[, which claim 11 depends from,] Strasser in an
analogous art" teaches garbage collection updates of non-volatile flash
memories. (Final Act. 14--15.) According to the Examiner:
It would have been obvious to one of ordinary skill in the art at
the time the invention was made to modify the teaching of
Cal[l] I Blankenship with the teaching of Strasser that
comprises garbage collection updates.
This modification would have been obvious to one of ordinary
skill in the art, at the time the invention was made, because one
of ordinary skill in the art would have recognized the need for
increased data reliability and efficiency in NVM memories
systems.
(Final Act. 15.) Appellants contend that it would not have been obvious to
combine the references because "the Examiner has not shown how one of
ordinary skill in the art would go about making this substantial modification
without rendering Call or Strasser unsuitable for their intended purposes, or
whether it could even be achieved at all." (App. Br. 17, emphasis added;
Reply 7-8.) The Examiner disagrees and responds that "as explained in the
final office action Call teaches metadata comprises partial parity data stripes
and Strasser is an obvious modification to comprise garbage collection
metadata." (Ans. 7.) In other words, the Examiner finds that it would have
been obvious to one of ordinary skill in the art to add Strasser' s teachings
regarding garbage collection metadata to Call' s teachings regarding
9
Appeal2015-007443
Application 13/567,945
metadata management. Appellants have not persuaded us that the Examiner
erred. We find that the Examiner provides sufficient articulated reasoning
having a rational underpinning, such that a person of ordinary skill in the art
would have been motivated to combine the teachings of Blankenship, Call,
and Strasser. (Id.) See KSR, 550 U.S. at 415, 418. Appellants' contention
that one of ordinary skill in the art would not be able to combine the
teachings of Blankenship, Call, and Strasser "without making substantial
modification" is mere attorney argument and a conclusory statement that is
unsupported by factual evidence. In re Geisler, 116 F.3d at 1470.
Appellants have not persuaded us that the Examiner erred and we agree with
the Examiner's finding regarding this claim.
Claim 13 recites that "the metadata manager is to provide to the
device a memory address of the memory cell circuitry at which the metadata
is stored; and the request includes the memory address." Appellants contend
that "Call fails to teach the provision of the memory address at which this
'metadata' is stored to an external device" and neither does Blankenship
"teach anything analogous to this 'metadata' stored in a 'spare area' in a
page in a corresponding stripe." (App. Br. 10.) The Examiner responds that
"Call in figure 1 teaches metadata stored in the external volatile memory ref.
(164)." (Ans. 7.) Appellants do not reply. Appellants have not persuaded
us that the Examiner erred and we agree with the Examiner's finding
regarding this claim.
For the foregoing reasons, we are not persuaded of Examiner error in
the rejection of claims 1, 7, 10, 11, and 13. Thus, we sustain the 35 U.S.C.
§ 103 rejections of claims 1, 7, 10, 11, and 13, as well as claims 2, 3, 6, 9,
12, and 14--40, which are not argued separately. (App. Br. 16.)
10
Appeal2015-007443
Application 13/567,945
Claims 4, 5, and 8
Claim 4 depends from claim 3 and further recites "the metadata
comprises a memory utilization metric associated with a specified memory
address range; and the at least one metadata management operation
comprises updating the memory utilization metric responsive to a match
between the specified memory address range and a memory address
identified by the memory access request." (Emphases added.) Appellants
contend that neither Call nor Blankenship teaches or suggests "a memory
utilization metric" much less one that is updated based on "a match between
the specified memory address range and a memory address identified by the
memory access request." (App. Br. 7-8; Reply 4.) The Examiner finds that
"Call in Figures 4A/B teaches measuring partial stripe (i.e. memory
utilization metric) to mark parity marking with metadata mis-match." (Ans.
5, emphasis in original; Final Act. 7.) We agree with Appellants that the
portions of Call cited by the Examiner do not teach or suggest the limitation
at issue. Specifically, we agree that "measuring partial stripe ... to mark
parity marking with metadata mis-match" is not "memory utilization metric"
nor is such a metric updated based on "a match between the specified
memory address range and a memory address identified by the memory
access request" as claimed.
Claim 5 depends from claim 3 and further recites "the metadata
comprises memory log information representing accessed memory locations;
and the at least one metadata management operation comprises updating the
memory log information with a memory address identified by the memory
access request." (Emphasis added.) The Examiner finds that Call teaches
this limitation because the controller uses the updated metadata "to track the
11
Appeal2015-007443
Application 13/567,945
locations of the valid pages." (Call, 6:21-30; Ans. 5-6.) Appellants
contend that "[ n ]owhere in Call is it disclosed or suggested that a memory
address is stored ... much less ... is updated 'with a memory address
identified by the memory access request.'" (App. Br. 8; Reply 5.) We agree
with Appellants that the portions of Call cited by the Examiner do not teach
or suggest the limitation at issue. Specifically, the portions of Call cited by
the Examiner do not teach or suggest "updating the memory log information
with a memory address identified by the memory access request" as claimed.
Claim 8 depends from claim 7 and further recites "the metadata
comprises address translation information; and the at least one metadata
management operation comprises accessing the address translation
information to translate the virtual address to the physical address."
Appellants contend that "the 'metadata' of Call [contains] valid page
information identifying which pages are valid for a corresponding stripe[,
which] is not the same as, or even equivalent to, address translation
information." (App. Br. 9.) The Examiner responds that "Call teaches
address translation each time the parity page is read, e.g.[,] in Figure 4B,
step ( 420) Call provides address translation to find out which pages are
valid." (Ans. 6; Final Act. 9.) We agree with Appellants that the portions of
Call cited by the Examiner do not teach or suggest the limitation at issue.
Specifically, step 420 of Figure 4B merely states "Read Metadata to Find
Out Which Pages Are Valid" and do not teach or suggest "translat[ing] the
virtual address to the physical address."
12
Appeal2015-007443
Application 13/567,945
For the foregoing reasons, we are persuaded of Examiner error in the
rejection of claims 4, 5, and 8, and do not sustain the 35 U.S.C. § 103
rejection of these claims. 2
DECISION
We affirm the decision of the Examiner to reject claims 1-3, 6, 7, and
9--40 under 35 U.S.C. § 103(a). We reverse the decision of the Examiner to
reject claims 4, 5, and 8 under 35 U.S.C. § 103(a).
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l )(iv).
AFFIRMED-IN-PART
2 Because we do not sustain the Examiner's rejection for the reasons
discussed herein, we need not address Appellants' other arguments relating
to these claims. See Beloit Corp. v. Valmet Oy, 742 F.2d 1421, 1423 (Fed.
Cir. 1984) (finding an administrative agency is at liberty to reach a decision
based on "a single dispositive issue").
13