Ex Parte Alimi et al

Patent Trial and Appeal Board

Apr 30, 2014

12115257 (P.T.A.B. Apr. 30, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
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.
12/115,257 05/05/2008 Richard A. Alimi YOR920070628US1 4758
59144 7590 04/30/2014
CAHN & SAMUELS, LLP
1100 17th STREET, NW
SUITE 401
WASHINGTON, DC 20036
EXAMINER
LIN, WEN TAI
ART UNIT PAPER NUMBER
2454
MAIL DATE DELIVERY MODE
04/30/2014 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 RICHARD A. ALIMI,
WILLIAM SWEENEY, and NIANJUN ZHOU
____________________

Appeal 2011-011838
Application 12/115,257
Technology Center 2400
____________________

Before: JOHN C. KERINS, JAMES P. CALVE, and
ANNETTE R. REIMERS, Administrative Patent Judges.

CALVE, Administrative Patent Judge.

DECISION ON APPEAL

STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134 from the final rejection of
claims 1-20. App. Br. 3. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM-IN-PART.
Appeal 2011-011838
Application 12/115,257

2
CLAIMED SUBJECT MATTER
Claims 1, 9, and 16 are independent. Claim 1 is reproduced below.
1. A system for distributing task assignments on a computer
network comprising:
a client grid having at least one server coupled to at least
one client node,
a plurality of client computers coupled to said client node
through a plurality of monitoring agents;
wherein each said monitoring agent collects data regarding
the available resources of a particular client computer and
periodically transmits said data to a grid server and said grid
server anticipates the available resources of each client computer
coupled to said client node based on previously received data
regarding the available resources of each client computer and
distributes grid tasks to each client computer of the grid based on
said previously received data regarding the available resources.

REJECTIONS
Claims 1-8 and 19 are rejected under 35 U.S.C. § 103(a) as being
unpatentable over Powers (US 2006/0075079 A1; iss. Apr. 6, 2006).
Claims 9-18 and 20 are rejected under 35 U.S.C. § 103(a) as being
unpatentable over Powers and Willen (US 7,058,949 B1; iss. Jun. 6, 2006).
ANALYSIS
Claims 1-8 and 19 unpatentable over Powers
The Examiner found that Powers discloses a system for distributing
task assignments substantially as claimed including a monitoring agent that
collects data regarding available resources of a particular client computer,
but does not teach that transmission of resource availability data to a grid
server is done periodically. Ans. 4. The Examiner also found that Powers
teaches the use of a metering agent to monitor agent activities and resources
being used, and an availability store that stores information about the current
Appeal 2011-011838
Application 12/115,257

3
status of all active computing resources. Ans. 4. The Examiner reasoned
that the availability store must be updated so the metering agent transmits
resource usage to the control server on a periodic basis and that it would
have been obvious to convey resource availability data to the grid server
periodically to allow the grid server to maintain updated resource
availability for allocation of resources. Ans. 4-5, 8. The Examiner also
found that Powers teaches a job manager that uses a status updater to update
the current status of active computing resources stored in an availability
store and the status updater must receive status updates of all of the active
computing resources. Ans. 9.
Appellants argue that Powers teaches a distributed processing system
that delegates the allocation and control of computing tasks to intelligent
agent applications that run on computing nodes, collects information about
those computing nodes, and uses that information to assess the ability of the
computing node to complete jobs that are posted on the server. App. Br. 8
(citing Powers, paras. [0033-0034]. Appellants also argue that each agent
manages execution of work units on its computing node and returns results
for a job rather than transmitting data regarding the available resources of a
particular client computer to a grid server that distributes grid tasks to each
client computer of the grid based on resource availability data received from
an agent. App. Br. 8. Appellants further argue that the meter agent monitors
the intelligent agent and measures the resources that the agent uses on the
system rather than providing periodic updates of resource availability. Id.
The Examiner has not established by a preponderance of evidence that
Powers discloses monitoring agents that collect data regarding the available
resources of a particular client computer and periodically transmit the data to
Appeal 2011-011838
Application 12/115,257

4
a grid server that distributes grid tasks to each client computer based on the
previously-received data regarding the available resources. See App. Br. 6,
8. Powers discloses intelligent agents that collect information for computing
nodes on which they run, use that information to assess the ability of the
computing node to complete jobs posted on a server, and manage execution
of the work units on each computing node. Powers, para. [0033]. Agents
measure the processor, main memory, network capacity, storage capacity,
and other attributes of the computing node on which it is installed. Powers,
para. [0033]. The Examiner has not explained how these measurements
track the available or idle resources of a client computer versus the total
resources of a client computer. Appellants disclose that a monitoring agent
320 collects the idle resources of a client computer and returns the collected
results to the grid server. Spec. 7, para. [0022]; see Spec. 3, para. [008].
This data relates to computing resources that are available to the grid such as
a client computer’s CPU availability, the available memory, the available
network bandwidth, the amount of time the client computer’s CPU dedicates
to grid tasks, the time interval that the resource is available to the grid, how
long a resource is idle, and general data as to when a resource is available.
Spec. 7, para. [0024]. This arrangement avoids duplicative tasking and
wasting time in continued attempts to assign a task to a client computer
whose resources are being used for other tasks. Spec. 3, para. [006].
Nor has the Examiner explained how a Meter Agent collects data for
available resources of a client computer. Powers discloses that a Meter
Agent monitors intelligent agents and hosted applications and measures the
resources that they use including the amount of CPU time used, peak CPU
used, amount of data read from or written to the hard disk, and number of
Appeal 2011-011838
Application 12/115,257

5
hard disk accesses. Powers, para. [0151]. Although Powers discloses an
availability store that stores information on the current status of all active
computing resources including new or reintroduced computing resources
(see Powers, para. [0058]), the Examiner has not explained how the current
status of active computing resources necessarily discloses the idle resources
of a computing resource that are available to perform grid tasks.
The Examiner also determined that although Powers does not use the
term “resource availability” it would have been obvious to a skilled artisan
that resource availability can be derived by subtracting a respective resource
usage from a total resource amount of its kind. Ans. 8-9. Even if resource
availability can be derived in this manner, that capability does not make it
obvious to modify Powers to do so. The Examiner has not explained how
the cited portions of Powers provide a rational underpinning for this change.
Powers discloses that information captured by the Meter Agent is sent to a
control server where it can be aggregated and used for informational, cost-
allocation, billing, price setting, storing, and to present results for computing
resources. Powers, para. [0151]. This disclosure suggests that Powers is
concerned with tracking computing resources that are used, rather than idle.
Powers also discloses that a status updater updates a current status of active
computing resources (para. [0058]), and the Examiner has not explained
how this disclosure necessarily indicates that the status information relates to
available computing resources or renders such a feature obvious. Ans. 9.
We do not sustain the rejection of claims 1-8 and 19.
Appeal 2011-011838
Application 12/115,257

6
Claims 9-18 and 20 unpatentable over Powers and Willen
Appellants argue claims 9-18 and 20 as a group. App. Br. 8-10. We
select claim 9 as representative. 37 C.F.R. § 41.37(c)(1)(vii) (2011). Claims
10-18 and 20 stand or fall with claim 9.
The Examiner found that Powers discloses a method of distributing
task assignments from a server to client computers on a grid, as recited in
claim 9, except for forming a resource probability distribution based on
historical computing resource data and employing a scheduling algorithm to
distribute grid tasks to the client computers using at least the probability
distribution. Ans. 7. The Examiner found that Willen teaches these features
by disclosing a task scheduler/dispatcher that uses a probability algorithm to
assign task resources based on usage history, which affects the probability
distribution of the resource. Ans. 7. The Examiner found that claim 9 does
not recite any particulars of the resource probability distribution nor does
Appellants’ Specification provide any definition or other disclosure relating
to the claimed resource probability distribution that distinguishes over the
combined teachings of Powers and Willen. Ans. 9-10. The Examiner also
found that Powers teaches the use of an agent core module 715 that predicts
the availability of a computing resource based on past patterns of usage that
form a probability and Willen teaches an example showing how a lottery
probability model is implemented for resource assignment. Ans. 10.
Appellants argue that Willen discloses a system supporting multiple
classes of tasks by allocating processor resources to classes according to a
probability model that teaches away from Appellants’ use of historical
computing resource availability data to anticipate the availability of grid
computing resources and efficiently distribute grid tasking based on the
Appeal 2011-011838
Application 12/115,257

7
resources anticipated availability. App. Br. 9-10. Appellants argue that the
randomness of a lottery-type draw is contrary to the concept of anticipating a
grid resource’s availability based on actual historical usage. App. Br. 10.
These arguments are not persuasive for the following reasons.
Appellants have not identified any disclosure in Willen that criticizes,
discredits, or discourages the claimed solution and thus Appellants have not
shown that Willen teaches away from the claimed subject matter. Willen
discloses that the probability of IP resources being used for various tasks is
calculated based on historical usage and this historical resource availability
data is used to assign resources for future tasks. Willen, col. 6, ll. 33-47.
Appellants have not identified a definition or other disclosure of probability
distribution in their Specification that distinguishes over Willen’s resource
probability distribution nor have Appellants pointed to any claimed feature
of the resource probability distribution that distinguishes over Willen. We
also agree with the Examiner that Powers teaches an agent core module 715
that predicts availability of computing resources based on patterns of usage
of that computing resource and that Willen and Powers render obvious this
claimed subject matter. We sustain the rejection of claims 9-18 and 20.
DECISION
We REVERSE the rejection of claims 1-8 and 19 and AFFIRM the
rejection of claims 9-18 and 20.
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-IN-PART
llw