Ex Parte Wu et al

Patent Trial and Appeal Board

Aug 20, 2013

11673915 (P.T.A.B. Aug. 20, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte YUGUANG WU and CHARLES J. ZHOU

___________

Appeal 2011-003851
Application 11/673,915
Technology Center 2400
____________

Before DAVID M. KOHUT, ERIC B. CHEN, and IRVIN E. BRANCH,
Administrative Patent Judges.

CHEN, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2011-003851
Application 11/673,915

2
This is an appeal under 35 U.S.C. § 134(a) from the final rejection of
claims 1-20, all the claims pending in the application. We have jurisdiction
under 35 U.S.C. § 6(b). We affirm.

STATEMENT OF THE CASE
Appellants’ invention relates to allocating physical memory in a
distributed, shared memory system and for maintaining interaction with the
memory using a reservation protocol. A processor node broadcasts a
memory request message to a first subset of nodes via a communication
network. A reservation protocol includes a four-way handshake between the
requesting processor node and a memory node that fulfills the request.
(Abstract.)
Claim 1 is exemplary, with disputed limitations in italics:
1. A system, comprising:
a processor node comprising a processor and a network
interface;
a plurality of memory nodes, wherein each memory node
comprises a memory and a network interface;
a communication network configured to interconnect said
processor node and said plurality of memory nodes via said network
interfaces according to a network topology, wherein each of said
memory nodes is removed from said processor node by a respective
number of network hops according to said network topology;
wherein said processor node is configured to:
broadcast a memory request message comprising a
unique identifier of a given memory request and a requested
memory size to a subset of said plurality of memory nodes;
send a confirmation message comprising the unique
identifier to a particular one of the subset of memory nodes in
Appeal 2011-003851
Application 11/673,915

3
response to receiving a reservation acknowledgment message
from said particular one of the memory nodes; and
access memory on the particular one of the memory
nodes in response to receiving a commit message from the
particular one of the memory nodes indicating that the memory
has been allocated for the given memory request.

REJECTION AT ISSUE
Claims 1-20 stand rejected under 35 U.S.C. § 103(a) as being obvious
over Richardson (U.S. Patent No. 6,249,802 B1; June 19, 2001) and
Yamauchi (U.S. Patent No. 5,649,102; July 15, 1997).

ANALYSIS
We are unpersuaded by Appellants’ arguments (App. Br. 11-12) that
the combination of Richardson and Yamauchi would not have rendered
obvious independent claim 1, which includes the limitation “broadcast a
memory request message . . . to a subset of said plurality of memory nodes.”
The Examiner found that the physical memory allocator of
Richardson that identifies a node set corresponds to the limitation “broadcast
a memory request message . . . to a subset of said plurality of memory
nodes.” (Ans. 4; Richardson, col. 5, ll. 16-30.) We agree with the
Examiner.
Richardson relates to “allocating physical memory in a distributed
shared memory (DSM) network.” (Abstract.) Richardson explains that the
DSM architecture includes “a plurality of physically distinct and separated
processing nodes each having one or more processors, input/output devices
and main memory that can be accessed by any of the processors.” (Col. 1,
ll. 20-25.) Figure 1A of Richardson illustrates an operating system 100 that
Appeal 2011-003851
Application 11/673,915

4
controls applications 102 and libraries 104 running on a distributed shared
memory (DSM) network 120. (Col. 5, ll. 8-12.) “Upon receiving a memory
configuration request, physical memory allocator 110 performs a search of
global geometry data and outputs a node set 125.” (Col. 5, ll. 27-29.) Thus,
Richardson teaches the limitation “broadcast a memory request message . . .
to a subset of said plurality of memory nodes.”
Appellants argue that “Richardson teaches a search for memory nodes
that takes place within the physical memory allocator of the operating
system” and that “[t]his is a completely different method for identifying
memory nodes that could meet a memory request than the four-message
memory allocation protocol of Appellants’ claimed invention.” (App.
Br. 12; see also Reply Br. 2-3.) However, the claim limitation “broadcast a
memory request message . . . to a subset of said plurality of memory nodes”
is broad enough to encompass the memory configuration request of
Richardson that identifies a node set 125 from a DSM network 120 to fulfill
the request. Furthermore, the Examiner cited to Yamauchi, rather than
Richardson, for teaching the limitations “send a confirmation message”,
“receiving a reservation acknowledgment message”, and “a commit
message.” (Ans. 5, 13.)
Thus, we agree with the Examiner that the combination of Richardson
and Yamauchi would have rendered obvious independent claim 1, which
includes the limitation “broadcast a memory request message . . . to a subset
of said plurality of memory nodes.”
We are further unpersuaded by Appellants’ arguments (App. Br. 13-
16; see also Reply Br. 4-7) that the combination of Richardson and
Yamauchi would not have rendered obvious independent claim 1, which
Appeal 2011-003851
Application 11/673,915

5
includes the limitations “send a confirmation message comprising the unique
identifier to a particular one of the subset of memory nodes in response to
receiving a reservation acknowledgment message from said particular one of
the memory nodes” and “access memory on the particular one of the
memory nodes in response to receiving a commit message from the
particular one of the memory nodes indicating that the memory has been
allocated for the given memory request.”
The Examiner acknowledged that Richardson does not disclose the
limitations “send a confirmation message comprising the unique identifier to
a particular one of the subset of memory nodes in response to receiving a
reservation acknowledgment message from said particular one of the
memory nodes” and “access memory on the particular one of the memory
nodes in response to receiving a commit message from the particular one of
the memory nodes indicating that the memory has been allocated for the
given memory request” (Ans. 4-5) and thus, relied upon Yamauchi for
teaching a distributed shared memory management system that includes
shared data management commands (Ans. 5, 13; Yamauchi, col. 25, ll. 30-
37). The Examiner concluded that “[i]t would have been obvious . . . to
combine Richardson and Yamauchi, by adding the reservation system of
Yamauchi to that of Richardson, because it . . . allows the computer to
operate more efficiently.” (Ans. 5.) We agree with the Examiner.
Yamauchi relates to “[a] distributed shared memory management
system for a distributed shared memory computer system . . . , each
computer having an independent address space and logically sharing data
physically distributed to a storage of each computer.” (Abstract.) Yamauchi
explains that “[w]hen a group is designated by the group joining
Appeal 2011-003851
Application 11/673,915

6
command 3001, the commands in the coherence control command memory
area 1701 are read and executed to enter necessary data items in the group
management table 2902.” (Col. 24, ll. 63-66.) In one embodiment, shared
data management commands 2007 include a coherence control request
command “ACQUIRE” 3301 (i.e., the claimed “a confirmation message”), a
coherence control acknowledge command “ACQUIRE_ACK” 3302 (i.e., the
claimed “reservation acknowledgment message), and a group joining
command “JOIN_GROUP” 3304 (i.e., the claimed “commit message).
(Col. 25, ll. 30-37.)
A person of ordinary skill in the art would have recognized that
incorporating the distributed shared memory management system of
Yamauchi, that includes shared data management commands, with the
physical memory allocator of Richardson would improve Richardson by
providing the ability to more effectively access a shared memory. See KSR
Int’l Co. v. Teleflex Inc., 550 U.S. 398, 417 (2007). Alternatively, the
combination of Richardson and Yamauchi is nothing more than
incorporating the known distributed shared memory management system of
Yamauchi with the known physical memory allocator of Richardson, to
yield predictable results. Id. at 416. Thus, we agree with the Examiner
(Ans. 5) that modifying Richardson to include the distributed shared
memory management system of Yamauchi would have been obvious.
Appellants argue that:
The two-message ACQUIRE/ACQUIRE ACK sequence for
requesting access to shared data as taught by Yamauchi (in
which an acquire message is sent by one computer and an
acquire_ack message is sent by another computer) cannot be
equivalent to the four-message sequence for requesting and
Appeal 2011-003851
Application 11/673,915

7
receiving an allocation of memory that is recited in Appellants’
claim 1.
(App. Br. 13-14; see also Reply Br. 4-5.) Again, the Examiner cited to
Richardson for teaching the limitation “broadcast a memory request
message” (Ans. 4) and the Examiner cited to Yamauchi, rather than
Richardson for teaching the limitations “send a confirmation message”,
“receiving a reservation acknowledgment message”, and “a commit
message” (Ans. 5, 13).
Appellants also argue that “there would have been no reason to
combine Richardson and Yamauchi in a manner that would result in
Appellants’ claimed invention.” (App. Br. 15; see also Reply Br. 6.)
However, as discussed previously, the combination of Richardson and
Yamauchi is based on the improvement of a similar device in the same way
as in the prior art, or in the alternative, by combining known elements to
achieve predictable results.
Thus, we agree with the Examiner that the combination of Richardson
and Yamauchi would have rendered obvious independent claim 1, which
includes the limitations “send a confirmation message comprising the unique
identifier to a particular one of the subset of memory nodes in response to
receiving a reservation acknowledgment message from said particular one of
the memory nodes” and “access memory on the particular one of the
memory nodes in response to receiving a commit message from the
particular one of the memory nodes indicating that the memory has been
allocated for the given memory request.”
Accordingly, we sustain the rejection of independent claim 1 under 35
U.S.C. § 103(a). Claims 2-11 depend from claim 1, and Appellants have not
presented any substantive arguments with respect to these claims.
Appeal 2011-003851
Application 11/673,915

8
Therefore, we sustain the rejection of claims 2-11 under 35 U.S.C. § 103(a),
for the same reasons discussed with respect to independent claim 1.
Independent claims 12, 17, and 19 recite limitations similar to those
discussed with respect to independent claim 1, and Appellants have not
presented any substantive arguments with respect to these claims. See In re
Lovin, 652 F.3d 1349, 1357 (Fed. Cir. 2011) (“[T]he Board reasonably
interpreted Rule 41.37 to require more substantive arguments in an appeal
brief than a mere recitation of the claim elements and a naked assertion that
the corresponding elements were not found in the prior art.”). We sustain
the rejection of claims 12, 17, and 19, as well as dependent claims 13-16, 18,
and 20 for the same reasons discussed with respect to claim 1.

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