Ex Parte Muralimanohar et al

Patent Trial and Appeal Board

Jul 1, 2016

13092912 (P.T.A.B. Jul. 1, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
13/092,912 04/23/2011 Naveen Muralimanohar
56436 7590 07/06/2016
Hewlett Packard Enterprise
3404 E. Harmony Road
Mail Stop 79
Fort Collins, CO 80528
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.
82275806 7665
EXAMINER
GEBRIL, MOHAMED M
ART UNIT PAPER NUMBER
2135
NOTIFICATION DATE DELIVERY MODE
07/06/2016 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):
hpe.ip.mail@hpe.com
mkraft@hpe.com
chris.mania@hpe.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte NA VEEN MURALIMANOHAR,
JICHUAN CHANG, PARTHASARATHY RANGANATHAN,
DOE HYUN YOON, and NORMAN PAUL JOUPPI
Appeal2014-009722
Application 13/092,912
Technology Center 2100
Before JASON V. MORGAN, MELISSA A. HAAPALA, and
NABEEL U. KHAN, Administrative Patent Judges.
KHAN, Administrative Patent Judge.
DECISION ON APPEAL
Appellants 1 appeal under 35 U.S.C. § 134(a) from the Final Rejection
of claims 1-4 and 8-22. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.
1 According to Appellants, the real party in interest is Hewlett-Packard
Development Company, LP, a wholly-owned affiliate of Hewlett-Packard
Company with HPQ Holdings, LLC, as the general or managing partner.
App. Br. 1.
Appeal2014-009722
Application 13/092,912
THE INVENTION
Appellants' invention is an "adaptive memory system" that "matches
static, dynamic, and dynamically-adjustable retention and resiliency
characteristics of portions of the physical memory devices with specified
retention and resiliency characteristics specified for the data storage
allocations." Abstr. Exemplary independent claim 1 is reproduced below.
1. An adaptive memory system comprising:
a number of physical-memory devices; and
one or more memory controllers to provide a logical
address space to which the physical-memory devices and
data-storage allocations are mapped, the data-storage allocations
containing stored data for respective entities executable in a
system,
wherein the one or more memory controllers comprise a
memory-management layer to map a given one of the
data-storage allocations to a given portion of the
physical-memory devices, the mapping comprising:
comparing retention and endurance characteristics
of respective portions of the physical-memory devices
with retention and endurance characteristics specified for
the given data-storage allocation, and
selecting, based on the comparing, the given portion
of the physical-memory devices to store data of the given
data-storage allocation.
REFERENCES AND REJECTIONS
Claim 1 stands rejected under 35 U.S.C. § 103(a) as obvious over Kai
(US 2002/0004882 Al; publ. Jan. 10, 2002) and Bronner (US 2013/0148437
Al; publ. June 13, 2013). Final Act. 2-6.
2
Appeal2014-009722
Application 13/092,912
Claims 2 and 8-10 stand rejected under 35 U.S.C. § 103(a) as obvious
over Kai, Bronner, and Wood (US 8,380,915 B2; iss. Feb. 19, 2013). Final
Act. 6-9.
Claim 3 stands rejected under 35 U.S.C. § 103(a) as obvious over Kai,
Bronner, Yano (US 8,209,471 B2; iss. June 26, 2012), and Tang (US
7,870,363 B2; iss. Jan. 11, 2011). Final Act. 9-12.
Claim 4 stands rejected under 35 U.S.C. § 103(a) as obvious over Kai,
Bronner, Yano, Tang, and Wood. Final Act. 12-13.
Claims 11-16, 20, and 22 stand rejected under 35 U.S.C. § 103(a) as
obvious over Kai and Wood. Final Act. 14-26.
Claims 17 and 18 stand rejected under 35 U.S.C. § 103(a) as obvious
over Kai, Wood, Yano, and Tang. Final Act. 26-29.
Claim 19 stands rejected under 35 U.S.C. § 103(a) as obvious over
Kai, Wood, Yano, Tang, and Kunimatsu (US 2010/0064111 Al; publ.
March 11, 2010). Final Act. 30-31.2
Claim 21 stands rejected under 35 U.S.C. § 103(a) as obvious over
Kai, Wood, and Tang. Final Act. 31-33.
ANALYSIS
Appellants present arguments for only the independent claims (claims
1, 11, and 14); each of which is rejected as obvious over Kai and one of
Bronner and Wood. See supra. Appellants contend each reference fails to
teach or suggest at-issue subject matter for which it is relied upon. See infra.
2 The Examiner omits Kai from claim 19 's statement of rejection. But, the
Examiner expressly relies upon Kai within the body of the rejection. See
Final Act. 30. Claim 19 is thus clearly rejected over Kai, Wood, Yano,
Tang, and Kunimatsu.
3
Appeal2014-009722
Application 13/092,912
We have reviewed the Examiner's rejections of the independent claims in
light of Appellants' arguments. We disagree with Appellants' arguments
and adopt as our own the findings set forth by the Final Office Action and
the Examiner's Answer and concur with the Examiner's conclusions. We
highlight the following only for emphasis.
Kai
Appellants argue Kai does not teach or suggest, as found by the
Examiner for claims 1, 11, and 14, comparing retention characteristics
respectively provided by portions of physical memory devices and desired
for a given data storage allocation. 3 App. Br. 6-8. The Reply Brief
summarizes the Examiner's findings and Appellants' arguments against Kai
as follows:
As explained in the Appeal Brief, ,-i [0085] of Kai refers to
a "storing table 9a [that] corresponds to the data retention time
storing means 9 . . . . " The data retention time storing means 9
stores data retention times of each of the rows in a DRAM. Kai,
,-i [0044]. ...
Importantly, note that there is no teaching or hint
anywhere in Kai that the data retention times (which apparently
was equated by the Examiner with "retention characteristics" in
claim 1) stored in the data retention time storing means 9 for each
row of a DRAM is compared with retention characteristics
specified for a given data-storage allocation (which according to
claim 1 contains stored data for a respective entity executable in
3 Appellants' Specification describes "retention" as follows: "The retention
time 728 for the memory cell is the time that elapses as the memory cell
drifts from the amorphous phase to an intermediate phase for which the data
state of the memory cell cannot be determined to a reasonable level of
certainty." Spec. 12, 11. 30-32.
4
Appeal2014-009722
Application 13/092,912
a system and is mapped by a memory management layer to a
given portion of the physical-memory devices).
Reply Br. 2; see also id. at 3 ("[Kai's] row of a DRAM having a longer data
retention time ... is used to store data that is to be allocated to memory ....
[And, Kai's] data with lifetimes that are close to each other are stored in the
same row").
We are unpersuaded by Appellants' arguments. Appellants do not
present persuasive argument or evidence that the Examiner errs in finding
Kai teaches comparing retention characteristics of portions of physical
memory devices with retention characteristics specified for a given data
storage allocation. See Reply Br. 4 (cursorily addressing Kai's fourth
embodiment); see also Ans. 24 (citing Kai's fourth embodiment description
(,-i 102)). Kai teaches selecting a row of physical memory for data allocation
by comparing the degrees of retention respectively provided by rows of
physical memory against a degree of importance desired for a given
data-storage allocation. Kai ,-i,-i 101-02; see also Ans. 24 ("[M]emory
allocation means 59 makes a reference to the data retention time of each row
stored in the data retention time storing means 9, disposes data on the
intermediate format 60 in accordance with the degree of importance of the
data[.]") Appellants fail to specifically rebut the Examiner's findings that
Kai's comparison of data row retention time to the importance of data
(retention characteristic of data storage allocation) teaches or suggests the
disputed limitation.
Bronner
Appellants argue Bronner does not teach or suggest, as found by the
Examiner for claim 1, a comparing of endurance characteristics respectively
5
Appeal2014-009722
Application 13/092,912
provided by portions of physical memory devices and desired for a given
data storage allocation.4 App. Br. 8-9. The Reply Brief summarizes the
Examiner's findings and Appellants' arguments against Bronner as follows:
Bronner refers to a memory endurance monitor "to record
a count of erase cycles performed by the device 1302." Bronner,
ii [0287]. "The monitor 1410 compares the recorded count
against a predefined count to determine whether to anneal the
device 1302." Id. As explained by Bronner, annealing a device
extends the utility of use-degraded components. Id., ii [0056].
Although reference is made to comparing the count of the
number of erase cycles to a predefined count (i.e. a threshold),
there is nothing in [Bronner5] that provides any teaching or hint
of comparing endurance characteristics of respective portions of
the physical-memory devices with endurance characteristics
specified for the given data-storage allocation (containing stored
data).
Reply Br. 5; see also App. Br. 9 ("Bronner refers to ... comparing the
recorded count of errors against a predefined number (i.e. a threshold) to
determine whether to anneal the device").
Essentially, Appellants argue Bronner presents no teaching or
suggestion to select a portion of physical memory for data allocation by
comparing degrees of endurance respectively provided by portions of
physical memory devices and desired for a given data storage allocation.
We are not persuaded of Examiner error, for each of two reasons.
4 Appellants' Specification describes "endurance" as follows: "The memory
cell represented by graph 702 is successfully cycled n times prior to failure,
so the cell exhibits an endurance ofn cycles." Spec. 12, 11. 15-17. The
meaning of "endurance" is not at issue.
5 Appellants state "nothing in Kai." The context of erase cycles and
endurance shows Appellants intended to state "nothing in Bronner."
6
Appeal2014-009722
Application 13/092,912
First, Appellants do not address the Examiner's finding that such a
suggestion arises from a combination of Kai's and Bronner' s teachings. See
infra; see also In re Keller, 642 F.2d 413, 426 (CCPA 1981) ("[O]ne cannot
show non-obviousness by attacking references individually where, as here,
the rejections are based on combinations of references").
Second, we agree with the Examiner's finding that Bronner teaches
endurance as a characteristic of memory reliability. See Ans. 4. We also
agree with the Examiner's finding that Kai teaches selecting a row of
physical memory for data allocation by comparing the degrees of retention-
another characteristic of memory reliability-respectively provided by rows
of physical memory against a degree of retention desired for a given
data-storage allocation. See Ans. 24; see also Kai ,-i,-i 101-02. We therefore
agree with the Examiner that it would have been obvious, in view of Kai, to
likewise use Bronner' s endurance characteristic as an indicator of memory
reliability; that is, to select a row of physical memory for data allocation by
comparing the degrees of endurance respectively provided by rows of
physical memory against a degree of endurance desired for a given
data-storage allocation.6
6 We further observe the likewise use of Bronner' s endurance characteristic
is further suggested by Kai's and Bronner' s inventions both determining and
storing representative values of their memory reliability characteristics
(respectively retention and endurance) for portions of physical memory. Kai
,-i 85; Bronner ,-i,-i 287-90; see also KSR Int 'l Co. v. Teleflex Inc., 550 U.S.
398, 417 (2007) ("simple substitution of one known element for another or
the mere application of a known technique to a piece of prior art ready for
the improvement"). We also observe the likewise use of Bronner' s
endurance characteristic is even further suggested by Kai's and Bronner's
inventions operating at a row granularity of physical memory. Kai ,-i 101;
Bronner ,-i 259.
7
Appeal2014-009722
Application 13/092,912
Wood
Appellants argue Wood does not teach or suggest, as found by the
Examiner for claims 11 and 14, a comparing of resiliency characteristics
provided by portions of physical memory devices and specified for a given
data storage allocation.7 App. Br. 11-12; Reply Br. 8-9. Appellants' Briefs
summarize three teachings of Wood (citing col. 28, 11. 49-54; col. 29, 11.
2-15) and assert each "provides no teaching or hint whatsoever of
comparing the specified retention and resiliency characteristics of the given
data-storage allocation with the retention and resiliency characteristics of the
portions of the physical-memory devices." Id. Appellants characterize one
teaching as concerning a measured reliability of physical memory;
particularly a counting of read errors. App. Br. 11; Reply Br. 8 ("comparing
the amount of errors to a default level"). And, Appellants characterize the
other two teachings as adjusting a read voltage (App. Br. 12; Reply Br. 8-9);
which we observe as, according to Wood, an action to decrease the
occurrence of read errors (Wood col. 28, 1. 61-col. 29, 1. 11).
Essentially, Appellants argue Wood presents no teaching or
motivation to select a portion of physical memory for data allocation by
7 Appellants' Specification describes "resiliency" as follows:
Reliability of memory devices, while depending on the materials
and architectures of the devices, may also be increased by
various post-manufacturing resiliency techniques. While failure
of memory cells may lead to unrecoverable data corruption in
memory devices, there are many different resiliency techniques
that can be employed to ameliorate up to threshold levels of
individual memory-cell failures.
Spec. 15, 11. 12-16. The meaning of "resiliency" is not at issue.
8
Appeal2014-009722
Application 13/092,912
comparing error rates provided by portions of physical memory devices and
error rates desired for a given data storage allocation. We are not persuaded
of Examiner error, for each of four reasons.
First, Appellants do not address the Examiner's finding that it would
have been obvious to use Wood's disclosed bias characteristic, not the
measured error rates, to select a portion of physical memory for data
allocation. Ans. 12, 15.
Second, Appellants argue Wood individually but do not address the
Examiner's finding that such a suggestion arises from a combination of
Kai's and Wood's cited teachings. See infra.
Third, we agree with the Examiner's finding that Wood teaches or
suggests the bias as a characteristic of memory reliability; particularly
teaching some memory cells are biased to more reliably store particular
types of data (having the same bias). See Ans. 12, 15 (citing Wood col. 29,
11. 12-24); see also Wood col. 18, 11. 57-60. We also agree with the
Examiner's finding that Kai teaches selecting a row of physical memory for
data allocation by comparing the degrees of retention-another
characteristic of memory reliability-respectively provided by rows of
physical memory against a degree of retention desired for a given
data-storage allocation. See Ans. 24. We therefore agree with the Examiner
that it would have been obvious, in view of Kai, to likewise use Wood's bias
characteristic as an indicator of memory reliability and resilience; that is, to
select a row of physical memory for data allocation by comparing the biases
9
Appeal2014-009722
Application 13/092,912
respectively provided by rows of physical memory against the bias of a
given data-storage allocation.8
Fourth, in addition to neglecting the Examiner's application of
Wood's bias to Kai's teachings, the arguments lack a persuasive premise.
As reflected by Appellants' arguments, Wood teaches the measured error
rates as representing a degree of reliability for a given data storage
allocation. Thus, we are not persuaded Kai and Woods fail to suggest
selecting a portion of physical memory for data allocation by comparing
measured error rates respectively provided for portions of physical memory
against a degree of reliability desired for a given data storage allocation.
Wood teaches the measured error rates as a characteristic of memory
reliability. Wood col. 1, 11. 48-54. Kai teaches selecting a row of physical
memory for data allocation by comparing the degrees of retention-another
characteristic of memory reliability-respectively provided by rows of
physical memory against a degree of retention desired for a given data
storage allocation. Kai ,-i 101. We therefore conclude it would have been
obvious, in view of Kai, to use Wood's measured error rates as an indicator
of memory reliability and resilience; that is, to select a row of physical
memory for data allocation by comparing the error rates respectively
provided by rows of physical memory against the degree of importance for a
given data storage allocation (indicating a permissible error rate relative to
other data allocations).
8 We observe such matching of a row's bias to a data allocation's bias
would, clearly, be provided in addition to Kai's matching of a row's
endurance to a data allocation's allowance for error. Kai ,-i 101.
10
Appeal2014-009722
Application 13/092,912
Conclusion
Appellants' arguments are not persuasive. Accordingly, we affirm the
Examiner's rejections of independent claims 1, 11, and 14. We also affirm
the Examiner's rejections of the remaining claims, each of which depend
from one of claims 1, 11, and 14, and which Appellants do not argue
separately.
DECISION
The Examiner's rejections of claims 1-4 and 8-22 under 35 U.S.C.
§ 103(a) are affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended. See 37 C.F.R. § l.136(a)(l )(iv).
AFFIRMED
11