Ex Parte Sauber et al

Patent Trial and Appeal Board

Jun 10, 2016

12791563 (P.T.A.B. Jun. 10, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
121791,563 06/01/2010
31625 7590 06/14/2016
BAKER BOTTS LLP,
PA TENT DEPARTMENT
98 SAN JACINTO BL VD., SUITE 1500
AUSTIN, TX 78701-4039
FIRST NAMED INVENTOR
William Sauber
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.
016295.4153 2941
EXAMINER
AYASH, MARW AN
ART UNIT PAPER NUMBER
2133
NOTIFICATION DATE DELIVERY MODE
06/14/2016 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):
DLAustinRecordsManagement@BakerBotts.com
tracy.perez@bakerbotts.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte WILLIAM SAUBER and STUART ALLEN BERKE
Appeal2015-001237
Application 12/791,563
Technology Center 2100
Before JOHN A. JEFFERY, BRADLEY W. BAUMEISTER, and
DENISE M. POTHIER, Administrative Patent Judges.
JEFFERY, Administrative Patent Judge.
DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner's
decision to reject claims 1-21. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.
Appeal2015-001237
Application 12/791,563
STATEMENT OF THE CASE
Appellants' invention reduces a computer memory's power
consumption from refresh operations and improves the memory's
performance. Spec. 6. For example, dynamic random access memory
(DRAM) stores each data bit in a separate capacitor. Id. at 4--5. Capacitors,
however, leak charge. Id. at 5. So unless the capacitor's charge is refreshed
periodically, the data bits will fade. Id. But refreshing DRAM cells
consumes power and limits useful work. Id. To improve performance and
power consumption, one embodiment tracks whether units are active or
inactive, and places the inactive units in self-refresh mode and active units in
command-based refresh mode. Id. at 6. Because the self-refresh mode does
not require a command, less power is consumed and other commands may
be issued. Id. at 15. Claim 1 is illustrative:
1. A method for improving performance and reducing
power consumption in a memory hardware system, comprising:
tracking, by a memory controller, whether individual
units of the memory system are active or inactive, wherein an
inactive individual unit is an individual unit that is in use, but
not currently being accessed;
placing, by the memory controller, inactive individual
units of the memory system in a self-refresh mode, such that the
inactive individual units self-refresh their contents; and
placing, by the memory controller, active individual units
of the memory system in a command-based refresh mode, such
that the active individual units are refreshed in response to a
received command to refresh their contents.
2
Appeal2015-001237
Application 12/791,563
THE REJECTION
The Examiner rejected claims 1-21under35 U.S.C. § 103(a) as
unpatentable over Foster1 (US 2007/0106860 Al; published May 10, 2007)
and Kim (US 2007/0253268 Al; published Nov. 1, 2007). Ans. 2--4.2
CONTENTIONS
The Examiner finds that Foster discloses every recited element of
claim 1 except for placing inactive units in a self-refresh mode, but relies on
Kim as teaching this feature in concluding that the claim would have been
obvious. Ans. 2-3. In particular, the Examiner finds that Foster's refresh
operation 42 is a command-based refresh for active units. Id. at 6 (citing
Foster Fig. 4). Under the proposed combination, the Examiner finds that
Kim's standby-mode DRAM cells are the recited inactive units. Ans. 3
(citing Kim i-f 3). Furthermore, the Examiner finds that Kim's DRAM cells
are self-refreshed. Ans. 3. Based on these collective teachings, the
Examiner concludes that it would have been obvious to put inactive units in
a self-refresh mode to retain the contents when in a low-power or standby
mode. Id.
Appellants argue that Foster does not refresh both inactive and active
units. App. Br. 8-9. In Appellants' view, Foster refreshes inactive rows, not
active ones. Id. at 8; see also id. at 9 (arguing Foster teaches away from
1 The Examiner notes that Foster incorporates by reference
Woo (US 2003/0028711 Al; published February 6, 2003 and referred to as
Woo71 l) and Woo (US 2003/0023825 Al; published January 30, 2003 and
referred to as W 00825). Ans. 2.
2 Throughout this opinion, we refer to (1) the Appeal Brief filed June 23,
2014 ("App. Br."); (2) the Examiner's Answer mailed August 29, 2014
("Ans."); and (3) the Reply Brief filed October 29, 2014 ("Reply Br.").
3
Appeal2015-001237
Application 12/791,563
refreshing active rows); Reply Br. 3--4 (citing Foster if 45). Appellants
further contend that the Woo references, incorporated by reference in Foster,
do not relate to active units. App. Br. 8-9.
Appellants contend that none of the cited references distinguish
between units placed in self-refresh mode and those placed in command-
based refresh mode. Id. at 9; Reply Br. 2-3. According to Appellants, the
recited inactive units are not currently being accessed, and active units are
"accessed more frequently." Reply Br. 3 (citing Spec. if 32).
Furthermore, Appellants argue that neither Foster, nor the references
that Foster incorporates by reference, describe whether step 42 relates to a
command-based or self-refresh mode. Reply Br. 4.
ISSUE
Under§ 103, has the Examiner erred in rejecting claim 1 by finding
that Foster and Kim collectively would have taught or suggested a refreshing
inactive and active units as recited?
ANALYSIS
On this record, we find no error in the Examiner's obviousness
rejection of claim 1.
Appellants do not dispute that Kim teaches inactive units as recited.
See App. Br. 8-9; Reply Br. 2--4. Rather, Appellants argue that Foster only
refreshes inactive units-namely, those units refreshed at operation 42. See,
e.g., Reply Br. 3 (citing Foster if 45).
But contrary to Appellants' characterization (see App. Br. 8-9;
Reply Br. 2--4), the Examiner finds that Foster's units refreshed at operation
4
Appeal2015-001237
Application 12/791,563
42 are the recited active units. Ans. 5 (discussing Foster Fig. 4 and
operation 42). And in the proposed combination, the Examiner finds that
Kim teaches inactive units and the recited self-refresh mode. Ans. 3. To the
finding regarding Kim, Appellants present no substantive rebuttal. See
App. Br. 8-9; Reply Br. 2--4. So in the absence of any rebuttal on this issue,
we are unpersuaded that the Examiner erred in finding that Kim places
inactive units in a self-refresh mode (Ans. 3).
For similar reasons, we are unpersuaded by the argument that the cited
references do not distinguish between units. See App. Br. 9; Reply Br. 2-3.
In particular, this argument amounts to an individual attack because it
ignores the Examiner's reliance on Foster in combination with Kim. See
App. Br. 9; Reply Br. 2-3. Such individual attacks do not show non-
obviousness where, as here, the rejection is based on the references'
collective teachings. See In re Merck & Co., Inc., 800 F.2d 1091, 1097
(Fed. Cir. 1986).
Accordingly, the only remaining question on appeal is whether Foster
places active units in a command-based refresh mode.
Notably, claim 1 recites explicitly that an inactive unit is one that is in
use, but not currently being accessed. Yet claim 1 recites no such definition
for active units. Nevertheless, the Specification explains that active units are
"accessed more frequently" than inactive units. See Spec. 15 :4--13. Accord
Reply Br. 3 (citing Spec. i-f 32).3 Therefore, Appellants' Specification
makes clear that active units need not be units that are currently being
3 Although Appellant cites paragraph 32 of the Specification, the distinction
between "active" and "inactive" areas of memory system 104 is discussed on
page 15, lines 6 to 11, of the originally-filed Specification.
5
Appeal2015-001237
Application 12/791,563
accessed, but, instead, can be units that are merely accessed more frequently
than those "not currently being accessed." So to the extent that Appellants
contend that active units must be those that are currently accessed, we
disagree: active units need only be accessed more frequently than inactive
units. And even this broad description is non-limiting, at least with respect
to active units. We reach this conclusion emphasizing the exemplary "e.g."
language used in connection with active units (Spec. 15: 10), and the
omission of this particular description of active units in the claim to so limit
its interpretation-an omission that contrasts starkly with the explicit
definition of inactive units in claim 1.
Given this breadth, we agree with the Examiner that the recited active
units cover Foster's units refreshed in operation 42 (Ans. 3). At operation
42, Foster refreshes units that (1) are in use, but (2) have not been refreshed
during the previous refresh cycle interval. Foster i-f 46, cited in Ans. 3. The
states of registers 24 and 36 reflect these two conditions. Foster i-f 45. In
particular, Foster's memory controller, or memory device, sets register 36 to
a digital logic high "1" if the row was recently refreshed. Id. i-f 46.
We are unpersuaded that register 36-----or more generally, whether a
unit was recently accessed---distinguishes Foster's units from the recited
active units. See App. Br. 8. Foster describes the memory unit as having
been "recently accessed" (Foster i-f 45), "recently used" (id. Fig. 4, operation
44), and "recently refreshed" (id. i-f 46) when register 36 is set high.
But as noted above, whether a unit is active relates to how frequently
it is accessed, not how recently it is accessed. Here, Appellants have not
shown that Foster's recent access reflects access frequency. Indeed, Foster's
register 36 does not count accesses. See e.g., Foster i-f 46. Rather, register
6
Appeal2015-001237
Application 12/791,563
36 toggles between binary states. See id. That is, the active units can be
those units refreshed in operation 42 (Ans. 3), which means that the unit was
not recently accessed, used, or refreshed (Foster Fig. 4, i-fi-1 45--46).
Appellants' arguments to the contrary, then, are not commensurate
with the scope of the limitation. So contrary to Appellants' arguments (see
App. Br. 8-9), we see nothing in Foster that precludes or teaches away from
refreshing active units given the breadth of the limitation. On this record,
the Examiner's finding that the units refreshed by Foster's operation 42 are
the recited active units (Ans. 3) is reasonable.
We also are unpersuaded by Appellants' argument that Foster lacks a
command-based refresh (Reply Br. 4). Specifically, Foster initiates an
action to refresh the memory rows. Foster i1 46, cited in Ans. 3. The
Examiner finds that Foster issues a command at refresh operation 42. Ans. 5.
Appellants merely argue that Foster's disclosure is "insufficient" to
determine whether these actions are a command-based refresh. Reply Br. 4.
But Appellants provide no further arguments to distinguish Foster's refresh
actions from the claimed invention. See id. On this record, we are
unpersuaded that the Examiner erred in finding that Foster discloses a
command-based refresh of the above-discussed active units (Ans. 3).
Accordingly, we sustain the Examiner's rejection of claim 1, and
claims 2-21, not argued separately with particularity. See App. Br. 9;
Reply Br. 4.
DECISION
The Examiner's decision rejecting claims 1-21 is affirmed.
7
Appeal2015-001237
Application 12/791,563
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
8