Ex Parte Bali et al

Patent Trial and Appeal Board

Mar 18, 2014

11098854 (P.T.A.B. Mar. 18, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte BAHRI B. BALI, GORDAN G. GREENLEE and
RICHARD E. WEINGARTEN
___________

Appeal 2011-006504
Application 11/098,854
Technology Center 2400
____________

Before CARLA M. KRIVAK, CARL W. WHITEHEAD JR., and
MICHAEL J. STRAUSS, Administrative Patent Judges.

WHITEHEAD, JR., Administrative Patent Judge.

DECISION ON APPEAL
Appeal 2011-006504
Application 11/098,854

2
STATEMENT OF THE CASE
Appellants are appealing the final rejection of claims 1, 2, 5-15, 18-
20, and 23-27. Appeal Brief 2. We have jurisdiction under 35 U.S.C. § 6(b)
(2012).
We affirm.

Introduction
“Appellants’ invention generally relates [to] a system and method of
monitoring servers in a network . . .” and for “routing requests from a client
to a server based on performance metrics at several layers of the network.”
Appeal Brief 10.

Illustrative Claim (Emphasis Added)
1. A method of processing queries in a network, comprising the steps of:
a computing platform including hardware measuring performance of
one or more servers;
the computing platform computing a performance metric for each of
the one or more servers;
the computing platform computing a virtual cluster performance
metric for one or more virtual clusters based on the computed performance
metric for each of the one or more servers belonging to the one or more
virtual clusters, wherein the virtual cluster performance metric includes an
average of response times from member servers over a predetermined
period of time; and
routing a request to one of the one or more servers based on the
Appeal 2011-006504
Application 11/098,854

3
computed performance metric to balance a workload among the one or more
servers.
Rejections on Appeal1
Claims 1, 2, and 5-10 stand rejected under 35 U.S.C. §103(a) as being
unpatentable over Freeman (U.S. Patent Number 6,922,724 B1; issued July
26, 2005) and Attanasio (U.S. Patent Number 5,918,017; issued June 29,
1999). Answer 4-7.
Claim 11 stands rejected under 35 U.S.C. §103(a) as being
unpatentable over Freeman, Attanasio, and Kommula (U.S. Patent Number
7,254,626 B1; issued August 7, 2007). Answer 7-8.
Claim 12 stands rejected under 35 U.S.C. §103(a) as being
unpatentable over Freeman, Attanasio, and Gourlay (U.S. Patent Number
7,349,969 B2; issued March 25, 2008). Answer 8-9.
Claims 13 and 23-27 stand rejected under 35 U.S.C. §103(a) as being
unpatentable over Freeman and Anerousis (U.S. Patent Number 6,760,775
B1; issued July 6, 2004). Answer 9-15.
Claims 14, 15 and 18-20 stand rejected under 35 U.S.C. §103(a) as
being unpatentable over Freeman, Attanasio and Anerousis. Answer 15-18.

ANALYSIS
We have reviewed the Examiner’s rejections in light of Appellants’
arguments that the Examiner has erred. We disagree with Appellants’
contentions. We concur with the findings and reasons set forth by the
Examiner in the action from which this appeal is taken and the reasons set

1 The 35 U.S.C. § 112, second paragraph rejection was withdrawn by the
Examiner. Answer 3.
Appeal 2011-006504
Application 11/098,854

4
forth by the Examiner in the Answer in response to Appellants’ Appeal
Brief. However, we highlight and address specific findings and arguments
for emphasis as follows.

Claims 1, 2 and 5-12
Appellants contend:

Neither Freeman nor Attanasio teaches the virtual cluster
performance metric includes an average of response times from
member servers over a predetermined period of time, as recited
in the claimed invention. Freeman does not disclose virtual
clusters, much less a virtual cluster performance metric.
Instead, Freeman only discloses a plurality of servers 180
grouped together in a server farm 110. Since Freeman does not
disclose virtual clusters or a virtual cluster performance metric,
it is arguably impossible for Freeman to disclose the virtual
cluster performance metric includes an average of response
times from member servers over a predetermined period of
time, as recited in the claimed invention.

Appeal Brief 11.

Appellants further contend:

Attanasio also fails to teach or suggest the virtual cluster
performance metric includes an average of response times from
member servers over a predetermined period of time, as recited
in the claimed invention. Attanasio discloses a system and
method for providing dynamically alterable computer clusters
for message routing. Attanasio teaches dividing an encapsulated
cluster into several virtual encapsulated clusters (VEC), and
dynamically distributing incoming connections within a VEC
based on current server load metrics and according to a
configurable policy. Attanasio performs routing based on load
metrics of individual servers within each VEC, but makes no
mention of an average of response times from member servers
Appeal 2011-006504
Application 11/098,854

5
of a VEC. The only “average” that Attanasio discloses is an
“average weight” (col. 8, lines 28-45). However, this average
weight is not an average of response times, much less an
average of response times from member servers over a
predetermined period of time. The weights in Attanasio are
associated with load metrics, not with average response time.

Appeal Brief 11.

The Examiner finds:
Attanasio et al teaches computing metrics for virtual clusters
which is a group of servers in a cluster and using the computed
metric to make a routing decision and the decision is based on
computed metrics along with administrator configurable
parameters such as time thresholds which are interpreted to be
calculated and stored in the system (column 2, line 57 - column
3, line 10 and column 3, lines 44-64). Freeman et al teaches a
metric used in a routing determination includes the latency of
the servers in a group. The latency or response times of the
servers in a group is used to manage the overall server and
network load and in doing so overall the latency metric would
be understood by one of ordinary skill in the art to have
occurred over a period of time and be the average response time
of the servers (column 38, lines 15-21 and lines 35-59).

Answer 18-19.
Appellants disagree with the Examiner’s findings:
[T]hat one of ordinary skill in the art would interpret Freeman's
disclosure of “latency” as an average of response times from
member servers over a predetermined period of time, as recited
in claim 1. Latency is not an average of response times. Instead,
latency is simply a response time. For example, Freeman
describes latency as “latency (e.g., response time)” at col. 51,
lines 11-13. Freeman does not say that latency is an average of
response times, but rather that it is just response time.

Appeal 2011-006504
Application 11/098,854

6
Reply Brief 2.
Attanasio discloses, “Mbuddy 610 can use four different classes of
metrics to compute a weights function for the executor: input metrics, host
metrics, service metrics, and user metrics.” Attanasio, column 7, lines 10-12.
Attanasio also discloses, “[T]he evaluation consists in producing appropriate
requests for each of the cluster host servers and measuring their answering
delays.” Id. at lines 48-50. The weights referred to in Attanasio include an
average of the response times since it uses the weights and the weights are
based upon metrics and the metrics may include an answering delay, which
is a response time of used to determine response time. Id. at 10-62.
Accordingly, Attanasio teaches or suggests the virtual cluster performance
metric includes an average of response times from member servers over a
predetermined period of time as recited in claim 1.
Appellants have not presented sufficient evidence or argument to
persuade us that averaging response time from member severs would have
been “uniquely challenging or difficult for one of ordinary skill in the art” or
“represented an unobvious step over the prior art.” See Leapfrog Enters.,
Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed. Cir. 2007) (citations
omitted). Therefore, we affirm the Examiner’s obviousness rejection of
claim 1, as well as dependent claims 2 and 5-10. We also affirm the
Examiner’s rejection of dependent claims 11 and 12 because they depend
from claim 1 and thus fall therewith.

Claims 13 and 23-27
Appellants contend:
Freeman does not disclose a virtual cluster. Since Freeman does
not disclose a virtual cluster, Freeman cannot reasonably be
Appeal 2011-006504
Application 11/098,854

7
construed as teaching computing a performance metric for a
virtual cluster and routing a request based on the performance
metric of the virtual cluster.

Anerousis also fails to teach a virtual cluster. The
passages of Anerousis identified by the Examiner describe an
“SLR cluster.” However, Anerousis's SLR cluster is not a
virtual cluster. Instead, the SLR is a “Service Level Router”
(Anerousis, col. 4, line 10). As shown in Anerousis FIG. 3,
reproduced below, the system level SLR cluster 310 functions
to route requests to a particular site, and the site-specific SLR
cluster 320, 325 routes requests to one of many servers at a
particular site. [Figure omitted].

However, the SLR cluster is not a virtual cluster. A
virtual cluster is a logical grouping of servers. See, e.g.,
Appellants’ specification, page 9, lines 6-10. Anerousis’s SLR
cluster is not a logical grouping of servers. Instead, Anerousis’s
SLR cluster is a router that routes requests to a grouping of
servers. Anerousis’s SLR cluster is associated with servers at a
site, but is not itself a grouping of servers. Therefore,
Anerousis’s SLR cluster is not a virtual cluster as recited in the
claimed invention.
Appeal Brief 16.

The Examiner finds:

Anerousis et al teaches calculating and using performance
metrics which include any criteria including but not limited to
load, cost, response, and proximity (column 11, lines 15-20).
The metric is then used to route a request for a virtual host to a
server (column 9, line 26 - column 10, line 33). The server is
interpreted as a virtual cluster due to hosting multiple virtual
host clients which receive client requests (column 6, line 66 -
column 7, line 7).

Answer 21.
Appellants further contend:
Appeal 2011-006504
Application 11/098,854

8
Claims 13 and 23 each positively recite: (i) a site, (ii) a
virtual cluster, and (iii) a server of a plurality of servers. The
Examiner, in equating the virtual cluster to a single server, is
effectively reading the “virtual cluster” limitation out of the
claims. That is, the Examiner's interpretation effectively
reduces the claim to only a site and a single server, without a
virtual cluster. However, this is improper since the plain
language of the claims requires all three elements, i.e., a site, a
virtual cluster, and a server of a plurality of servers.

For example, claim 13 recites computing respective
performance metrics for a site, a virtual cluster, and a server of
a plurality of servers. Use of the term “respective” means that
there are three performance metrics, i.e., one for each one of the
site, the virtual cluster, and the single server of the plurality of
servers. However, the Examiner's interpretation would merge
the virtual cluster into the single server, thus resulting in only
two performance metrics.

Moreover, Appellants submit that one of ordinary skill in
the art would not construe a “virtual cluster” as a single server.
Instead, as noted in the AB at page 16, a virtual cluster is a
logical grouping of a plurality of servers.

Reply Brief 5-6.
The argued features (virtual clusters and performance metrics) merely
pertain to statements of intended use. Claim 13 recites, in the body of the
claim, “a computing platform including hardware computing respective
performance metrics for a site, a virtual cluster, and a server of a plurality of
servers” and claim 23 recites, in the preamble, “[a] computer program
product comprising a computer usable storage medium having readable
program code embodied in the medium, the computer product includes at
least one component.” “An intended use or purpose usually will not limit
the scope of the claim because such statements usually do no more than
Appeal 2011-006504
Application 11/098,854

9
define a context in which the invention operates.” Boehringer Ingelheim
Vetmedica, Inc. v. Schering-Plough Corp., 320 F.3d 1339, 1345 (Fed. Cir.
2003). Although “[s]uch statements often . . . appear in the claim’s
preamble,” In re Stencel, 828 F.2d 751, 754 (Fed. Cir. 1987), a statement of
intended use or purpose can appear elsewhere in a claim. Id.
Therefore, we sustain the Examiner’s obviousness rejections of claims
13 and 23 for the reasons the Examiner articulated in the Answer (9-11 and
21) as discussed above.2

Appellants also argue:
Claim 24 recites virtual server clusters. Neither Freeman
nor Anerousis discloses virtual server clusters. The Examiner
admits that Freeman does not disclose virtual server clusters.
The Examiner asserts that Anerousis describes a virtual server
cluster at the SLR cluster. However, Anerousis’s SLR cluster is

2 In the event of further prosecution, we leave it to the Examiner to consider
if independent claim 23 should also be rejected under 35 U.S.C. § 101.
Signals are not patentable eligible subject matter under § 101. In re Nuijten,
500 F.3d 1346, 1357 (Fed. Cir. 2007). See also United States Patent and
Trademark Office (USPTO), Subject Matter Eligibility of Computer
Readable Media, 1351 Off. Gaz. Pat. Office 212 (Feb. 23, 2010), the policy
statement by former USPTO Director David J. Kappos, as published in the
Official Gazette of the USPTO. The aforementioned USPTO policy
statement guides:
The broadest reasonable interpretation of a claim drawn to a
computer readable medium (also called machine readable
medium and other such variations) typically covers forms of
non-transitory tangible media and transitory propagating signals
per se in view of the ordinary and customary meaning of
computer readable media, particularly when the specification is
silent. (Emphasis added).

Appeal 2011-006504
Application 11/098,854

10
not a virtual server cluster. Instead, as described supra, the SLR
cluster is a router. Since Anerousis does not disclose a virtual
server cluster, it is impossible for Anerousis to teach a
performance metric for a virtual server cluster. Therefore,
Anerousis cannot be relied for teaching generating a virtual
cluster performance metric for each of a plurality of virtual
server clusters associated with the site, as recited in claim 24.

Appeal Brief 19.

The Examiner finds:
Anerousis et al teaches using performance metrics to route a
request for a virtual host to a server (column 9, line 26 - column
10, line 33). The server is interpreted as a virtual cluster due to
hosting multiple virtual host clients which receive client
requests (column 6, line 66 - column 7, line 7).

Answer 21.

Appellants further argue:
In the Response to Arguments section at page 21 of the
Answer, the Examiner asserts the same arguments as noted
above with respect to claims 13 and 23. That is, the Examiner
asserts that a single server of Anerousis is interpreted as the
virtual cluster due to the description of Anerousis at col. 6, line
66 through col. 7, line 7. Appellants disagree for the same
reasons noted above.

Reply Brief 7.
Claim 24 recites, in the preamble, “[a] system for balancing workload,
comprising at least one hardware component configured to.” The recitation
“configured to” is nonfunctional claim language, which is not entitled to any
Appeal 2011-006504
Application 11/098,854

11
patentable weight. See Superior Indus., Inc. v. Masaba, Inc., 2014 WL
163046 at *5 (Fed. Cir. Jan. 16, 2014) (Rader, C.J., concurring) stating:
[A] system claim generally covers what the system is, not what
the system does. Hewlett-Packard Co. v. Bausch & Lomb Inc.,
909 F.2d 1464, 1468 (Fed. Cir. 1990); see also Roberts v. Ryer,
91 U.S. 150, 157 . . . (1875) (“The inventor of a machine is
entitled to the benefit of all the uses to which it can be put, no
matter whether he had conceived the idea of the use or not.”).
Thus, it is usually improper to construe non-functional claim
terms in system claims in a way that makes infringement or
validity turn on their function. Paragon Solutions, LLC v.
Timex Corp., 566 F.3d 1075, 1091 (Fed. Cir. 2009).
At best, the claim language “configured to” is a statement of intended
use that merely indicates the capability or intent to perform the action of
receiving the acknowledgment. Our reviewing court has held that a
statement of intended use in an apparatus claim cannot distinguish over a
prior art apparatus that discloses all the recited structural limitations and is
capable of performing the recited function. See In re Schreiber, 128 F.3d
1473, 1477 (Fed. Cir. 1997). We also note that “[a]n intended use or
purpose usually will not limit the scope of the claim because such statements
usually do no more than define a context in which the invention operates.”
See Boehringer Ingelheim Vetmedica, Inc. v. Schering-Plough Corp., 320
F.3d 1339, 1345 (Fed. Cir. 2003). Although “[s]uch statements often . . .
appear in the claim's preamble,” a statement of intended use or purpose can
appear elsewhere in a claim. In re Stencel, 828 F.2d 751, 754 (Fed. Cir.
1987).
Therefore, we do not find Appellants’ arguments persuasive for the
reasons the Examiner articulated in the Answer 21, and as discussed above.
Subsequently, we sustain the Examiner’s obviousness rejection of claim 24.
Appeal 2011-006504
Application 11/098,854

12
We also sustain the Examiner’s obviousness rejection of claim 25, which
depends upon claim 24, for the same reasons discussed above.
Appellants argue claim 26 is distinguishable over the combination of
Freeman and Anerousis:
Since Freeman and Anerousis fail to teach a virtual
server cluster, it follows that they do not teach: (i) determine
virtual cluster performance metrics of the plurality of virtual
server clusters; (ii) after the selecting the selected site, select
one of the plurality of virtual server clusters for handling the
request based on the virtual cluster performance metrics; and
(iii) after the selecting the selected virtual server cluster, select
one of the plurality of servers of the selected virtual server
cluster for handling the request based on the server
performance metrics, which are also recited in claim 26 and
which are defined in terms of a virtual server cluster.

Appeal Brief 22.

The Examiner finds:
Anerousis et al teaches using performance metrics to route a
request for a virtual host to a server (column 9, line 26 - column
10, line 33). The server is interpreted as a virtual cluster due to
hosting multiple virtual host clients which receive client
requests (column 6, line 66 - column 7, line 7).

Answer 22.

At the outset, we note claim 26 is directed to a “method to provide a
service for establishing a computing infrastructure.” However, the body of
claim 26 is directed to “a computing platform having at least one hardware
component that operates to” perform several steps. Subsequently, the
limitations in the body of claim 26 are not directed to the method for
providing a service as stated in the preamble, the limitations are actually
Appeal 2011-006504
Application 11/098,854

13
directed to the operations of the computing platform having a hardware
component. Thus, as an initial matter of claim construction, we conclude it
is unclear whether the “method to provide a service” or a “computing
platform having at least one hardware component” “operates to” perform the
particular operations recited in the body of claim 26.3
Thus, Appellants’ arguments are unpersuasive because the plain
language of claim 26 only requires a method to provide a service for
establishing a computing infrastructure by providing a computing platform
having at least one hardware component and subsequently fails to
distinguish over the cited prior art as discussed above. Therefore, we sustain
the Examiner’s obviousness rejection of claim 26 and claim 27, which
depends upon claim 26, for the reasons discussed above.

Claims 14, 15 and 18-20
Appellants submit that the applied art does not teach the
combination of features recited in independent claim 14. For

3 Our reviewing court guides that a claim directed to a system and a method
for using that system is indefinite. See IPXL Holdings, L.L.C. v.
Amazon.com, Inc., 430 F.3d 1377, 1384 (Fed. Cir. 2005); see also Ex parte
Miyazaki, 89 USPQ2d 1207, 1211 (BPAI 2008) (precedential) (“[I]f a claim
is amenable to two or more plausible claim constructions, the USPTO is
justified in requiring the applicant to more precisely define the metes and
bounds of the claimed invention by holding the claim . . . indefinite.”). In
the event of further prosecution, we leave it to the Examiner to consider
whether at least independent claim 1 (and associated dependent claims)
should also be rejected under 35 U.S.C. § 112, second paragraph, as being
indefinite. While the Board is authorized to reject claims under 37 C.F.R. §
41.50(b), no inference should be drawn when the Board elects not to do so.
See Manual of Patent Examining Procedure (MPEP) § 1213.02.

Appeal 2011-006504
Application 11/098,854

14
example, as discussed above, Freeman, Attanasio, and
Anerousis fail to teach the virtual cluster performance metric
includes an average of response times from member servers
over a predetermined period of time. To the contrary, neither
Freeman nor Anerousis teaches an average of response times of
anything, much less an average of response times from member
servers of a virtual cluster over a predetermined period of time.
And, as discussed above with respect to claim 1, Attanasio does
not inherently teach the virtual cluster performance metric
includes an average of response times from member servers
over a predetermined period of time. The Examiner has
provided no basis in fact or technical reasoning to support the
assertion that Attanasio inherently teaches this feature, and
instead is resorting to unfounded speculation as to how the
Attanasio system might be operating.

Appeal Brief 24.

The Examiner finds:
Freeman teaches the latency or response times of the servers in
a group is used to manage the overall server and network load
and in doing so overall the latency metric would be understood
by one of ordinary skill in the art to have occurred over a period
of time and be the average response time of the servers (column
38, lines 15-21 and lines 35-59). Anerousis et al teaches
performance metrics are used and calculated for a cluster and a
server and are both used to make a routing decision to each of a
cluster and a server and the metrics include any criteria
including but not limited to load, cost, response, and proximity
(column 9, line 26 - column 10, line 33 and column 11, lines
15-20).

Answer 23-24.

Again, at the outset, we note claim 14 is directed to a “system for
routing requests in a network, comprising at least one hardware component
to” perform the recited steps or operations. However, the list of operations
Appeal 2011-006504
Application 11/098,854

15
in the body of claim 14 is recited in the manner similar to the steps of a
method claim.4 Thus as an initial matter of claim construction, we conclude
that it is unclear whether a “system” or “hardware component” is performing
the particular operations or steps recited in the body of claim 14 Further, the
list of operations in the body of claim 14appears to be statements of intended
use for the “at least hardware component.” Therefore, we do not find
Appellants’ arguments to be persuasive because the plain claim language
only requires a system having a hardware component capable of performing
the steps or operations recited in the body of claim 26 and subsequently fails
to distinguish over the cited prior art as discussed above.
We sustain the Examiner’s obviousness rejection of independent claim 14,
as well as dependent claims 15 and 18-20, not separately argued.

DECISION
The Examiner’s obviousness rejections of claims 1, 2, 5-15, 18-20,
and 23-27 are 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). See 37
C.F.R. § 41.50(f).

AFFIRMED

tj

4 See previous footnote.