Ex Parte Dawson et al

Patent Trial and Appeal Board

Aug 19, 2013

11243658 (P.T.A.B. Aug. 19, 2013)

UNITED STATES PATENT AND TRADEMARKOFFICE
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
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
11/243,658 10/04/2005 Christopher J. Dawson END920040203US1 9950
7590 08/19/2013
Whitham, Curtis, & Christofferson. P.C.
Suite 340
11491 Sunset Hills Road
Reston, VA 20190
EXAMINER
CHANG, JULIAN
ART UNIT PAPER NUMBER
2452
MAIL DATE DELIVERY MODE
08/19/2013 PAPER
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.
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte CHRISTOPHER J. DAWSON, RICK A. HAMILTON II, JOSHY
JOSEPH, and JAMES W. SEAMAN
____________

Appeal 2011-000495
Application 11/243,658
Technology Center 2400
____________

Before BRUCE R. WINSOR, LARRY J. HUME, and
GREGG I. ANDERSON, Administrative Patent Judges.

ANDERSON, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s final
rejection of claims 1-16. We have jurisdiction under 35 U.S.C. § 6(b). We
affirm.

STATEMENT OF THE CASE
Appellants’ invention relates to monitoring, evaluating, archiving and
publishing to related grid modules comprehensive historical statistics which
allow for increasingly accurate and efficient management of a computing
grid as a whole. See generally Spec. 9, ll. 1-9. Claim 1is illustrative:
Appeal 2011-000495
Application 11/243,658

2

1. A method of managing a grid computing environment including
steps of:
performing job portions of a data processing request using respective
resources of said grid computing environment;
periodically capturing and storing as data records, in correspondence
with a characteristic of a respective job portion, a snapshot of current
instantaneous operations of individual application environments in said grid
computing environment from respective resources performing said job
portions of said data processing requests in content-addressable storage, said
data records corresponding to performance of respective ones of said
respective resources while processing respective ones of said job portions,
said performance being monitored in correspondence with characteristics of
respective ones of said job portions, such that any performance data or other
parameter of said data records for respective ones of said resources may be
retrieved based on any other performance data or parameter of said data
records or logical combination thereof with arbitrarily fine granularity;
retrieving selected data records in accordance with data stored in said
data records; and
processing data retrieved in said retrieving step to produce processed
performance data as historical performance statistics.

THE REJECTIONS
1. The Examiner rejected claims 1-3 under 35 U.S.C. § 103(a) as
being unpatentable over Lee (Run-Time Prediction of Parallel Applications
on Shared Environments, 2003), Yue (Global Backfilling Scheduling in
Multiclusters, 2004), and applicant-admitted prior art (“AAPA”). Ans. 4-7.1
2. The Examiner rejected claims 6 and 7 under 35 U.S.C. § 103(a) as
being unpatentable over Lee, Yue, and Horvitz (US 7,519,564 B2; Apr. 14,
2009, filed Jun. 30, 2005). Ans. 7-8.

1 Throughout this opinion, we refer to (1) the Appeal Brief filed June 7,
2010 (“App. Br.”); (2) the Examiner’s Answer mailed July 19, 2010
(“Ans.”); and (3) the Reply Brief filed September 20, 2010 (“Reply Br.”).
Appeal 2011-000495
Application 11/243,658

3

3. The Examiner rejected claim 4 under 35 U.S.C. § 103(a) as being
unpatentable over Lee, Yue, AAPA, and Krsul (VMPlants: Providing and
Managing Virtual Machine Execution Environments for Grid Computing,
2004). Ans. 7.
4. The Examiner rejected claims 9 and 10 under 35 U.S.C. § 103(a)
as being unpatentable over Lee, Yue, AAPA, and Bjornson (US
2004/0056908 A1; published Mar. 25, 2004). Ans. 8.
5. The Examiner rejected claim 5 under 35 U.S.C. § 103(a) as being
unpatentable over Lee, Yue, AAPA, and Blythe (The Role of Planning in
Grid Computing, 2003). Ans. 8-9.
6. The Examiner rejected claim 8 under 35 U.S.C. § 103(a) as being
unpatentable over Lee, Yue, AAPA, and Appleby (Oceano - SLA Based
Management of a Computing Utility, 2001). Ans. 9-10.
7. The Examiner rejected claims 11-16 under 35 U.S.C. § 103(a) as
being unpatentable over Lee, Yue, and Buyya (Nimrod/G: An Architecture
for a Resource Management and Scheduling System in a Global
Computational Grid, 2000). Ans. 10-13.

ISSUES
(1) Under § 103, has the Examiner erred in rejecting claim 1 by
finding that Lee, Yue, and AAPA collectively would have taught or
suggested
periodically capturing and storing as data records, in correspondence
with a characteristic of a respective job portion, a snapshot of current
instantaneous operations of individual application environments in
said grid computing environment from respective resources
performing said job portions of said data processing request in
Appeal 2011-000495
Application 11/243,658

4

content-addressable storage, said data records corresponding to
performance?

(2) Under § 103, has the Examiner erred in rejecting claim 6 by
finding that Lee, Yue, and Horvitz collectively would have taught or
suggested “periodically storing data records including storing data records in
a data repository”?
(3) Under § 103, has the Examiner erred in rejecting claim 4 by
finding that Lee, Yue, AAPA, and Krsul collectively would have taught or
suggested “processing a request for proposal based on historical
performance analysis”?
(4) Under § 103, has the Examiner erred in rejecting claim 9 by
finding that Lee, Yue, AAPA, and Bjornson collectively would have taught
or suggested: “determining if portions of said data processing request are
application specific,” as recited in claim 9; or “interrogating the grid or
interrogating an application environment of the grid,” as recited in claim 10?
(5) Under § 103, has the Examiner erred in rejecting claim 5 by
finding that Lee, Yue, AAPA, and Blythe collectively would have taught or
suggested “determining a likelihood that a job run time can be performed
within an associated historical performance statistic”?
(6) Under § 103, has the Examiner erred in rejecting claim 8 by
finding that Lee, Yue, AAPA, and Appleby collectively would have taught
or suggested basing the division of claim 7 on “how said grid computing
environment is dispersed over multiple security domains”?
(7) Under § 103, has the Examiner erred in rejecting claim 11 by
finding that Lee, Yue, and Buyya collectively would have taught or
Appeal 2011-000495
Application 11/243,658

5

suggested “means for managing said grid computing environment based on
said processed performance data”?

ANALYSIS
Rejection of Claims 1-3 under 35 U.S.C. § 103
The majority of Appellants’ briefing on claim 1 is directed toward
alleged deficiencies in the teachings of Lee. We will initially focus on what
the Examiner finds was not taught by Lee but is taught by Yue and AAPA,
i.e., the “snapshot” limitation. Ans. 6. We therefore review the record
regarding this disputed limitation of claim 1 which recites, in pertinent part,
“a snapshot of current instantaneous operations of individual application
environments in said grid computing environment from respective resources
performing said job portions of said data processing requests in content-
addressable storage, said data records corresponding to performance.”
(Emphasis added).
The Examiner finds Yue teaches periodically capturing a “snapshot”
of current operations. Ans. 6, 16 (citing Yue, §3). The snapshot includes
individual application environments and “job portions” in a grid computing
environment. Id. The Examiner cites to Yue’s teaching that each cluster of
the grid “sends a snapshot of its own scheduling information, which includes
resource utilization profile, running jobs profile and waiting jobs queue
profile at fixed interval.” Id. The Examiner finds that “content-addressable
storage” is AAPA because Appellants failed to adequately traverse Official
Notice of that position in responding to a prior office action. Ans. 6.
Appellants start off by generally alleging the snapshot limitation has
not, prior to the claimed invention, been available. App. Br. 22. Capturing
Appeal 2011-000495
Application 11/243,658

6

and storing snapshots of performance of the components of the grid
computing system enhance performance, e.g., “run-time prediction as well
as RFP generation, pricing, discounts and the like.” App. Br. 27. More
substantively, Appellants argue that while Yue does use the word
”snapshot,” the usage is in the context of scheduling information. App. Br.
28. Yue’s “scheduling information” is not the capture of performance data
as claimed. App. Br. 29, Reply Br. 7. Appellants contend the Examiner’s
position with respect to the asserted AAPA that “it would be obvious to one
of ordinary skin in the art at the time of applicant’s invention to employ
content addressable storage in order to improve performance.” App. Br. 20.
Appellants agree with the statement, but argue that just because content
addressable storage was previously known does not establish that the
unexpected functions or capabilities claimed were also known. Id.
Appellants’ arguments do not persuade us of any error by the
Examiner in finding the disputed limitation present in the cited prior art.
Both Lee and Yue are in the field of grid computing. Lee, §1; Yue, §1.
Appellants do not dispute the combination.
As the Examiner finds, Yue’s teaching of sending a snapshot of
scheduling information is “performance data.” Ans. 16. The Examiner’s
citation to Yue also includes reference to a “resource utilization profile,”
which would also include information in the nature of “performance data.”
Id. “Scheduling information” is described in Yue as a scheduler structure,
which is “about scheduling events across geographically distributed clusters
at gird [sic].” Yue, §3. This is further described as an “issue of monitoring
[the] grid.” Id. Similarly, Appellants’ Specification describes “performance
data” in the context of managing the grid and is “collected on the grid.”
Appeal 2011-000495
Application 11/243,658

7

Spec. 5, ll. 14-17, ll. 26-27. No specific definition is given in the
Specification that would preclude “scheduling information” or a “resource
utilization profile” information from falling within the broadest reasonable
interpretation of “performance data.”
As to the monitoring by snapshots of the “job portions” of the grid
computing environment, the Examiner finds that the performing job
portions” limitation of claim 1 is taught by Lee. Ans. 5 (citing Lee, §2).
Specifically, Lee teaches a recording data to predict run times when “each
application run finishes.” Appellants merely deny Lee teaches the
limitation. App. Br. 28. In the context of the disputed limitation, we note
that section 3 of Yue cited by the Examiner further teaches taking the
snapshot in the context of a “jobs profile” and a “waiting jobs queue.” The
snapshot is taken at a “fixed interval.” Stated differently, performance
information is recorded during operation of the job and not just at the
completion of the application run.
The addition of “content-addressable storage,” which Appellants
admit is known in the art, is not a patentable difference over the prior art.
We understand Appellants’ argument on this issue to be that the claim
includes other functions that are not AAPA. This was not disputed by the
Examiner in finding those functions in the prior art other than AAPA. We
agree with the Examiner.
Beyond the snapshot limitation, Appellants specifically argue three
other limitations of claim 1 as not being present in Lee. App. Br. 28. We
partially addressed the first limitation above, “performing job portions.” We
are not persuaded of error by what is simply a denial that Lee teaches the
limitation. See In re Lovin, 652 F.3d 1349, 1357 (Fed. Cir. 2011). The
Appeal 2011-000495
Application 11/243,658

8

Examiner cites to section 2 of Lee for the other two limitations. Ans. 5.
Again, Appellants’ denial that the limitations are present is insufficient for
us to find any error.
We have fully considered Appellants’ briefs and are unable to discern
any persuasive additional argument for patentability of claim 1. We sustain
the Examiner’s rejection of claim 1 for the reasons set forth above. Other
than the conclusory contention that their limitations are not present, claims 2
and 3 are not separately argued, and the rejection of those claims is likewise
sustained.

Rejection of Claims 6 and 7 under 35 U.S.C. § 103
Claim 6 depends on claim 1 and includes the additional step of
“periodically storing data records including storing data records in a data
repository.”
The Examiner relies on Horvitz to teach storing historical data in one
or more data stores. Ans. 7 (citing Horvitz, col. 13, ll. 1-11). As to whether
Horvitz is reasonably pertinent to the particular problem addressed by the
claimed invention, the Examiner finds Horvitz is directed towards predictive
modeling based on historical data. Ans. 16-17 (citing Horvitz, Fig. 1).
Appellants contend Horvitz is not related to quantitative prediction
and is not in a grid environment. App. Br. 30. Appellants do not contend
Horvitz is non-analogous art but rather irrelevant. Reply Br. 8.
The Examiner correctly cited Horvitz as analogous art and the
assertion was unrebutted. Therefore, Appellants’ attack on Horvitz’ alleged
lack of relevance is not persuasive. In re Keller, 642 F.2d 413, 426 (CCPA
1981) (“[O]ne cannot show non-obviousness by attacking references
Appeal 2011-000495
Application 11/243,658

9

individually where, as here, the rejections are based on combinations of
references.”). Additionally, Horvitz does address predictive analysis based
on historical data. Horvitz, col. 1, ll. 42-45. Though not used as a basis for
rejection by the Examiner, periodically storing data is a step that would have
been taken by the person of ordinary skill in the art. KSR Int’l Co. v.
Teleflex Inc., 550 U.S. 398, 418 (2007).
We sustain the Examiner’s rejection of claim 6. Claim 7 depends on
claim 6 and is not separately argued, and we also sustain the rejection to
claim 7.
Rejection of Claim 4 under 35 U.S.C. § 103
Claim 4 depends on claim 1 and adds the additional step of
“processing a request for proposal based on historical performance
analysis.”
The Examiner finds the additional step in Krsul. Ans. 8. Krsul
teaches providing a cost estimate. Id. (citing Krsul, § 3.2). Additionally,
Krsul’s teaching of cost-bidding to achieve resource utilization and load
balancing maps to the recited “request for proposal” language. Ans. 18
(citing Krsul, § 3.4). The Examiner finds Krsul is in the same field of
endeavor as the claimed invention, grid computing. Ans. 17.
Appellants assert Krsul is directed to configuring a grid architecture
for a virtual machine (VM) and is not relevant to run time prediction or other
elements of the claimed invention. App. Br. 30. Appellants also contend the
cost estimate taught by Krsul is not a “request for proposal” as claimed. Id.
Appellants’ arguments do not persuade us of any error. There is no
rebuttal of the combination of Krsul with other art by Appellants. Reply Br.
8-9. Again, Krsul is attacked individually, and not in the combination. The
Appeal 2011-000495
Application 11/243,658

10

term “request for proposal” is given no special meaning in the Specification
and, giving the term its broadest reasonable interpretation, we agree with the
Examiner that “cost-bidding,” as taught by Krsul, meets the limitation.
We therefore sustain the rejection of claim 4.

Rejection of Claims 9 and 10 under 35 U.S.C. § 103
Claim 9 depends on claim 1 and includes the additional step of
“determining if portions of said data processing request are application
specific.”
The Examiner finds the step taught in Bjornson. Ans. 19. Bjornson
teaches application-specific scheduling and scheduling logic. Id. (citing
Bjornson, ¶¶ [0009], [0015]). Appellants discuss paragraph 9 of Bjornson,
but do not explain why the teaching there is not applicable. Reply Br. 11.
Paragraph 9 of Bjornson states: “The key issues in dataflow systems
are application integration/interoperability (including data conversion and
data flow among the applications in the dataflows) and performance
improvement by means of sophisticated application-specific scheduling.”
(Emphasis added). The Examiner finds the additional step of claim 9 is
present in Bjornson based on this citation and we agree.
Claim 10 depends on claim 9 and adds the further step of
“interrogating the grid or interrogating an application environment of the
grid” based on the result obtained from the determining step of claim 10.
As noted above, the Examiner relies on paragraph 15 of Bjornson.
Appellants restate paragraph 15 and allege it does not teach the step. Reply
Br. 11-12.
Appeal 2011-000495
Application 11/243,658

11

Paragraph 15 of Bjornson teaches “[s]cheduling logic for selecting
which worker should perform a given task is coded in the taskbag
abstraction layer.” The Examiner mapped this language to the interrogating
step under review. We are not persuaded by Appellants that the Examiner’s
finding is not rationally based.
We sustain the rejections to claims 9 and 10.

Rejection of Claim 5 under 35 U.S.C. § 103
Claim 5 depends on claim 1 and includes the additional step of
“determining a likelihood that a job run time can be performed within an
associated historical performance statistic.”
The Examiner finds Blythe teaches generating a distribution of
possible runtimes. Ans. 9 (citing Blythe, § Plan Quality, p. 161). Blythe’s
teaching of distributing possible runtimes can reasonably be interpreted as a
historical performance statistic. Ans. 18.
Appellants merely restate what Blythe teaches, and contend those
teachings do not meet the claim limitation. App. Br. 32, Reply Br. 9.
Appellants do not contest the Examiner’s interpretation with anything
other than a denial that the cited language of Blythe does not meet the
limitation. The argument is not persuasive. See Lovin, 652 F.3d at 1357.
The rejection of claim 5 is sustained.

Appeal 2011-000495
Application 11/243,658

12

Rejection of Claim 8 under 35 U.S.C. § 103
Claim 8 depends on claim 7, which depends on claim 6, and includes
the further step of basing the division of the data repository on “how said
grid computing environment is dispersed over multiple security domains.”
The Examiner relies on Appleby to teach that management and
monitoring agents may reside on customer assigned Dolphin and Whale
servers, each of which are associated with a customer domain. Ans. 9-10
(citing Appleby, §1, p. 857). The Examiner concludes that Appleby teaches
customers’ databases “reside on one of a plurality of distributed fixed whale
servers, wherein each fixed whale server resides on a different security
domain.” Ans. 18.
Appellants acknowledge Appleby discloses distribution of servers
over domains so they can be reassigned. App. Br. 33. Appellants deny that
teaching meets the additional step claimed. Id. Appellants, again, argue the
teaching of Appleby is irrelevant. Reply Br. 10.
Appellants have not persuasively argued why the additional recited
step, dispersing the computing environment over “multiple security
domains,” is not met by the cited portion of Appleby. Appellants do not
contest that Appleby teaches multiple servers associated with different
domains.
We sustain the rejection to claim 8.

Appeal 2011-000495
Application 11/243,658

13

Rejection of Claims 11-16 under 35 U.S.C. § 103
Claim 11 is an independent claim to a system. It substantially tracks
the limitations of claim 1.2 Claim 11 recites, in pertinent part, “means for
managing said grid computing environment based on said processed
performance data.”
The rejection made by the Examiner also tracks the rejection made in
connection with claim 1. Ans. 10-11. As to the disputed limitation, the
Examiner relies on Buyya. Id. (citing Buyya, Scheduler, p. 285).
The arguments made by Appellants are a restatement of arguments
made in connection with claim 1. App. Br. 35. As to the disputed
limitation, Appellants acknowledge that a scheduler is “one aspect” of
managing said grid computing environment. Id. It is also argued, in the
Reply Brief, that Lee is inapplicable because data is collected at the end of a
run, and not in snapshots as per claim 11. Reply Br. 12.
Again, we are unpersuaded by Appellants’ arguments. Appellants
assert claim 11 is patentable over Buyya because of “unexpected and
improved functionalities.” App. Br. 36. There is no evidence in the record
before us of any such secondary considerations, nor is there any compelling
argument supporting this assertion.

2 Claims 11-16 recite limitations in means-plus-function form as permitted
by 35 U.S.C. § 112, 6th paragraph. In the event of further prosecution of
claims 1-16, or claims in similar form, we leave to the Examiner to ascertain
whether Appellants’ Specification discloses sufficient structure for the
means-plus-function limitations to render the limitations definite under
35 U.S.C. § 112, 2nd paragraph. See MPEP § 2181 II.B (8th ed. 2001; rev.
Aug. 2012); see also Aristocrat Techs. Australia Pty Ltd. v. Int’l Game
Tech., 521 F.3d 1328, 1338 (Fed. Cir. 2008).
Appeal 2011-000495
Application 11/243,658

14

We addressed the snapshot limitation in connection with claim 1. The
argument made was not presented in connection with claim 1, but does not
alter our opinion. Yue teaches “periodically” taking snapshots as previously
discussed.
We sustain the rejection of claim 11.
Claims 12-16 are argued based on their dependency on claim 11 and
Appellants contention that the cited art does not teach the recited limitations.
App. Br. 36-37, Reply Br. 14-15. This argument is not persuasive. To the
extent Appellants may be advancing additional arguments, they are
addressed below.
The Examiner rejects claim 12 on Buyya for teaching a plurality of
“modules for performing respective grid environment management functions
including at least one of cost estimation,” as recited in claim 12. Ans. 12
(citing Buyya, Resource Cost, p. 286, APIs, p. 287). The Examiner notes
that Appellants’ Specification discloses that each module is preferably
embodied as a software “object.” Id. (citing Spec. 15). The Examiner maps
Buyya’s teaching of “Resource Cost” to the recited “cost estimation.” Ans.
21 (citing Buyya, p. 286).
As to claim 12, Appellants argue Buyya is related to low level
protocols and APIs, which do not meet the recited limitation. App. Br. 36-
37. Further, Appellants allege resource cost, admittedly taught by Buyya, is
not “cost estimation.” Reply Br. 15. We are not persuaded that the broadest
reasonable interpretation of “cost estimation’ does not encompass “resource
cost.” Accordingly, we agree with the Examiner’s findings.
With regard to claim 13, the Examiner finds Buyya teaches a
Scheduler that manages jobs by selecting appropriate resources. Ans. 21
Appeal 2011-000495
Application 11/243,658

15

(citing Buyya, p. 285). Appellants argue Buyya’s Scheduler does not meet
the function of the “means for managing said grid computing environment”
recited in claim 13. App. Br. 36-37. A “Scheduler” is a “means for
managing,” as the Examiner finds.
The Examiner finds, as to claim 14, which recites that the process of
claim 11 includes “an accounting and statistics management module,”
Buyya’s resource discovery algorithm that keeps track of resources status
information maps to the limitation. Ans. 21 (citing Buyya, p. 285).
The Examiner finds, as to claim 15, which recites that claim 14 further
includes a “grid workload agent.” Claim 15 further requires the agent
receives processed performance data. The Examiner finds Buyya’s resource
selection algorithm uses processed data to select resources. Ans. 21 (citing
Buyya, p. 285).
In summary as to claims 14 and 15, the Examiner finds Buyya teaches
scheduling based on historical information. Ans. 21 (citing Buyya, p. 286).
Buyya’s Scheduler would require an “accounting and statistics module” for
remembering the historical information, and a “grid workload agent” that
performs scheduling based on the historical information. Id.
Regarding claim 16, which recites that claim 15 further includes
functional modules associated with the workload agent, the Examiner finds
the “functional modules associated with said grid workload agent” is a broad
limitation that is met by any programming in the scheduler (i.e., workload
agent) of Buyya. Ans. 21.
Appellants’ argument as to claims 14-16 is a conclusory denial that
the recited limitations are not met. Reply Br. 15. As has been repeatedly
Appeal 2011-000495
Application 11/243,658

16

stated here, such an argument is not persuasive. See Lovin, 652 F.3d at
1357.
The Examiner’s findings regarding claims 12-16 are rationally based
and not persuasively rebutted. We decline to find error on this record with
respect to claims 12-16. We sustain the rejection of claims 12-16.

CONCLUSION
On this record, the Examiner did not err in rejecting claims 1-16 under
§ 103.

ORDER
The Examiner’s decision rejecting claims 1-16 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

rwk