Ex Parte Grover et al

Patent Trial and Appeal Board

Sep 25, 2012

10821832 (P.T.A.B. Sep. 25, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte RAJIV K. GROVER and BANU AMALKANTI
____________

Appeal 2010-004979
Application 10/821,832
Technology Center 2100
____________

Before THOMAS S. HAHN, DENISE M. POTHIER, and
BARBARA A. BENOIT, Administrative Patent Judges.

BENOIT, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-004979
Application 10/821,832

2
STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s
rejection of claims 1-26. We have jurisdiction under 35 U.S.C. § 6(b). We
affirm-in-part.
Invention
Appellants’ invention relates to a technique for implementing device
loading in storage networks. See generally Abstract; Spec. ¶¶ 0005-06.
Claim 1 is reproduced below with key disputed limitations emphasized:
1. A computer program product including computer-readable storage
with a computer program, the computer program executing a computer
process on a computer system, the computer process:
identifying a plurality of storage devices to be configured in a storage
network;
identifying a number of host port Logical Unit Numbers (LUNs)
which are configured on each of the storage devices;
identifying a number of host port connections to the storage devices;
and
for each host port connection, determining actual loading of
input/output (IO) jobs for each of the storage devices based at least in part
on a queue depth for each of the host port LUNs.
The Examiner relies on the following as evidence of unpatentability:
Nahum US 2004/0078599 A1 Apr. 22, 2004
(filed Feb. 27, 2002)
Cherian US 7,225,242 B2 May 29, 2007
(filed Jan. 26, 2001)
JON TATE ET AL., IBM SAN SURVIVAL GUIDE 160-62 (2003), available at
http://www.redbooks.ibm.com/redbooks/pdfs/sg246143.pdf.

Appeal 2010-004979
Application 10/821,832

3
The Rejections
1. The Examiner rejected claims 1, 2, 4, 6, 7, 9, 11-14, 16-21, 23, 24,
and 26 under 35 U.S.C. § 103(a) as unpatentable over Cherian in view of
Tate. Ans. 3-9.1
2. The Examiner rejected claims 3, 5, 8, 10, 15, 22, and 25 under 35
U.S.C. § 103(a) as unpatentable over Cherian in view of Tate and Nahum.
Ans. 9-10.

OBVIOUSNESS REJECTION OVER CHERIAN IN VIEW OF TATE
CLAIMS 1, 2, 6, 7, 20, 21, 23, 24, AND 26
Regarding independent claim 1, the Examiner finds that Cherian
discloses every recited feature except for the queue depth for each of the
host port LUNs. Ans. 3-4. The Examiner also discusses Cherian’s teaching
that a server execution throttle is the maximum number of IO commands
that a server can have outstanding and is typically controlled by a
configuration setting in the host bus adapter (HBA) device driver. Ans. 4.
The Examiner then cites Tate as teaching a LUN-queue-depth parameter for
the HBA that is used to limit the number of outstanding commands per
LUN. Ans. 4. The Examiner, in concluding that the claim would have been
obvious, finds that Tate’s LUN queue depth is the same as the server
execution throttle taught by Cherian. Ans. 5.
Appellants argue that Cherian does not teach actual loading of IO jobs
for storage devices because execution throttle levels and potential command

1 Throughout this opinion, we refer to the Appeal Brief filed July 29, 2009;
the Examiner’s Answer mailed November 12, 2009; and the Reply Brief
filed January 9, 2010.
Appeal 2010-004979
Application 10/821,832

4
throughput of the servers is not the same as the actual loading of the storage
devices. App. Br. 6-7; Reply Br. 3-4.
Issue
Under § 103, has the Examiner erred in rejecting claim 1 by finding
that Cherian and Tate collectively would have taught or suggested
determining actual loading of IO jobs for each of the storage devices based
at least in part on a queue depth for each of the host port LUNs?
Analysis
Based on the record before us, we find no error in the Examiner’s
rejection of claim 1, which requires actual loading of IO jobs for storage
devices based at least in part on a queue depth for the host port LUNs.
Appellants have not persuaded us of error in the Examiner’s finding that
Cherian teaches determining actual loading of IO jobs for each of the storage
devices based at least in part on execution throttle level of each host. Ans. 4
(citing Cherian, col. 5, ll. 13-39).
Although Appellants correctly note that the Specification does not
expressly define “actual” (Reply Br. 3), nor by extension “actual loading” or
“maximum actual loading,” we are not persuaded by Appellants’ assertion,
based on a commonly understood meaning of actual, that Cherian’s “server
execution throttle” and “the maximum number of I/O commands” merely
exist potentially and, as such, cannot be interpreted as the claimed actual
loading. Reply Br. 3-4 (citing Merriam-Webster’s Online Dictionary
definition of actual as “existing in act and not merely potentially”).
Appellants’ Specification discusses device loading and IO job management
in a storage network using a variety of terms, including “actual loading,”
“device loading,” and, even, “maximum actual IO loading.” See generally
Appeal 2010-004979
Application 10/821,832

5
Spec. ¶¶ 0026-32. Appellants also assert that the Specification is
“inconsistent” in its usage of the term “actual loading” in paragraph 0028
where Appellants assert that “actual loading” in the equation means “device
loading.” Reply Br. 4. Appellants point to paragraph 0029 as disclosing the
correct distinction between “device loading” and “actual loading.” Id.
(quoting Spec. ¶ 0029). Appellants’ Specification in paragraph 0029 states:
“[T]his storage network may be designed . . . so long as the maximum actual
IO loading is less than the acceptable device loading” (emphasis added).
Thus, Appellants point to the maximum actual IO loading as actual loading.
Id.
Accordingly, we are not persuaded that the broadest reasonable
construction of “determining actual loading” cannot include determining an
actual loading for IO jobs for storage devices where the actual loading is a
maximum actual IO loading, which is consistent with the Specification.
Cherian teaches such an actual loading (e.g., summed execution throttle
value) of the jobs based on at least in part on a queue depth for each host
port LUNs. See Ans. 4 (citing Cherian, col. 5, ll. 13-39). Further, the
Specification’s discussion about determining a maximum actual IO loading
undercuts Appellants’ assertion that actual loading cannot mean execution
throttle levels and potential command throughput because actual means
“existing in act and not merely potentially.” Reply Br. 3. Maximum actual
IO loading would seem to suffer similar alleged deficiencies of being
“merely existing potentially” as server execution throttle and potential
command throughput do, according to Appellants. See Reply Br. 4. Yet
Appellants point to paragraph 0029 discussing maximum actual IO loading
as properly distinguishing device loading and actual loading. Id.
Appeal 2010-004979
Application 10/821,832

6
For the foregoing reasons, Appellants have not persuaded us of error
in the rejection of independent claim 1. Although dependent claim 2 was
separately argued, the arguments are similar to those made in connection
with claim 1. App. Br. 7-8; Reply Br. 4. We are not persuaded of error in
the rejection of claim 2 for the above reasons.
Independent claim 6 recites, in pertinent part, for each host port
connection, determining actual loading for each of the storage devices based
at least in part on a queue depth for each of the host port connections.
Claims 6 and 7 were not separately argued with particularity. See App. Br.
7; see generally Reply Br. 3-10. Therefore, for the same reasons described
above with respect to claims 1 and 2, we are not persuaded that the
Examiner erred in rejecting claim 6 or its dependent claim 7.
Independent claim 20 recites, in pertinent part, for each of a plurality
of host port connections to the at least one storage device, determining actual
loading of the at least one storage device based at least in part on a queue
depth for each of the host port connections so that the number of IO jobs
being issued by a host do not exceed the queue depth of a service queue.
Claims 20, 21, 23, 24, and 26 were not separately argued with particularity.
See App. Br. 11-12; see generally Reply Br. 3-10. Therefore, for the reasons
described above with respect to claim 1, we are not persuaded that the
Examiner erred in rejecting claim 20 or its dependent claims 21, 23, 24, and
26.
CLAIMS 4 AND 9
Regarding claims 4 and 9, which depend from independent claims 1
and 6 respectively, the Examiner finds that Cherian discloses using a loading
factor to determine if the actual loading for each of the storage devices
Appeal 2010-004979
Application 10/821,832

7
exceeds a maximum loading. Ans. 5 (citing Cherian, col. 5, ll. 21-44), 13-
14; Reply Br. 8-9. Appellants merely repeat the additional limitation recited
by claim 4, quote a portion of the reference cited by the Examiner, and
conclude that “[s]umming the execution throttles for various servers in
Cherian is not the same as using a loading factor to determine if the actual
loading for each of the storage devices exceeds a maximum loading.”
Appeal Br. 9 (emphases in original); see also Reply Br. 6. The explanation
Appellants provide is merely that the portion of Cherian’s disclosure cited
by the Examiner is better mapped to the Specification’s determining
maximum queue depth. Reply Br. 6. We are not persuaded.
We agree with the Examiner’s explanation that Cherian discloses the
loading factor in Equation 1 or 2 in the cited portion of column 5, not merely
in the summations of the execution throttles as alleged by Appellants.
Ans. 13. As recognized by the Examiner (id.), Cherian discloses summing
“the execution throttle levels or command throughput of each network
server” and a “verification test is performed to determine whether the
summed execution throttle value exceeds the command queue depth or
command throughput of the associated storage controller.” Cherian, col. 5,
ll. 13-21. Also, giving the recited “loading factor” its broadest reasonable
construction, we note that Appellants’ Specification supports an embodiment
where the loading factor is one. See Spec. ¶ 0028 (discussing that Actual
loading = (1*2*8*10) = 160, where the first value of 1, while not explained,
seems to represent the loading factor). As such, Appellants’ disclosure
undercuts the arguments (App. Br. 8-9; Reply Br. 5-6) objecting to the
Examiner’s finding that Cherian teaches the claimed loading factor.

Appeal 2010-004979
Application 10/821,832

8
Additionally, on this record, Appellants have not met their burden to
rebut the Examiner’s prima facie case of obviousness because Appellants
merely present assertions without providing factual evidence. See App. Br.
9. Mere arguments unsupported by factual evidence are entitled to little
probative value. Cf. In re Geisler, 116 F.3d 1465, 1470 (Fed. Cir. 1997).
Therefore, we are not persuaded that the Examiner’s rejection of
claims 4 and 9 is erroneous.
CLAIMS 11-14, 16, AND 17
Independent claim 11 recites, in pertinent part, automatically
determining actual loading for the storage device based on at least in part on
a queue depth for each host port LUN. For the rejection of independent
claim 11 and its dependent claims 12-14, 16, and 17, Appellants refer to the
previous arguments of claim 1. App. Br. 9-10; Reply Br. 7-8.
Appellants also contend that the Examiner erred in not according
appropriate patentable weight to the preamble – namely, a method for
providing an IO flow control mechanism in a storage network comprising
various steps. App. Br. 9-10; Reply Br. 7-8. In particular, Appellants argue
that the Examiner did not evaluate whether the purpose or intended use
recited in the preamble results in a structural difference between the claimed
invention and the prior art. App. Br. 10 (citing MPEP § 2111.02 (Effect of
Preamble)); Reply Br. 7.
The Examiner, however, indicated that the body of the claim supports
the preamble and so the result would be no different had the preamble been
given additional patentable weight. Ans. 14. Moreover, Appellants have not
provided factual evidence of any structural differences between the purpose
or intended use recited in claim 11 and the applied references to show
Appeal 2010-004979
Application 10/821,832

9
Examiner error. Mere arguments unsupported by factual evidence are
entitled to little probative value. Cf. Geisler, 116 F.3d at 1470.
Therefore for these reasons, and for the reasons described above with
respect to claim 1, we are not persuaded of error in the Examiner’s rejection
of independent claim 11 or its dependent claims 12-14, 16, and 17.
CLAIMS 18 AND 19
Regarding claim 18, which depends from independent claim 11, the
Examiner finds that Cherian and Tate collectively do not teach that the
maximum loading for the storage device is based on a loading factor for test
environments. Ans. 8. The Examiner concludes that it would have been
obvious, based on Cherian’s disclosure of the maximum loading in the
system for the production environment, to test the loading of the storage
devices prior to production to ensure that the system would perform with
minimum error. Id. Appellants challenge the Examiner’s conclusion as
impermissible hindsight. App. Br. 11.
The Examiner’s explanation that it would have been obvious to test
the storage devices does not provide a rational underpinning for “maximum
loading for the storage device is based on a loading factor for test
environments,” as required by claim 18 (emphasis added). While KSR Int’l
Co. v. Teleflex, Inc., 550 U.S. 398, 418 (2007) indicates that a precise
teaching directed to the claim language is not necessary in an obviousness
rejection, the Supreme Court also held that an Examiner must articulate
some reasoning with some rational underpinning to support a legal
conclusion of obviousness. Absent impermissible hindsight, we find the
Examiner has not demonstrated that Cherian and Tate collectively teach or
Appeal 2010-004979
Application 10/821,832

10
suggest the maximum loading for the storage device is based on a loading
factor for test environments.
For the foregoing reasons, Appellants have persuaded us of error in
the rejection of claim 18 and dependent claim 19 for similar reasons.

OBVIOUSNESS REJECTION OVER CHERIAN
IN VIEW OF TATE AND NAHUM
CLAIM 3
Claim 3 recites, inter alia, determining actual loading for each of the
storage devices based at least in part on a number of LUN security groups in
the storage network. In concluding that claim 3 is obvious, the Examiner
relies on Nahum’s disclosure of a system in which LUN security groups are
used. Ans. 9-10 (citing Nahum, ¶ 0018). Appellants correctly point out that
the cited portion of Nahum’s disclosure merely describe a security procedure
for authenticating each host, which is distinct from the claim language. App.
Br. 13; Reply Br. 10. The plain meaning of the claim requires determining
actual loading based on the number of LUN security groups – that is, the
claim requires taking action based on the number of LUN security groups.
We are therefore persuaded by Appellants’ argument that the mere
disclosure of a security procedure for authenticating each host does not teach
or suggest determining actual loading for storage devices based at least in
part on a number of LUN security groups in the storage network.
Based on the record before us, we are persuaded of error in the
Examiner’s rejection of claim 3, which requires “determining actual loading
for each of the storage devices based at least in part on a number of LUN
security groups in the storage network” (emphasis added).
Appeal 2010-004979
Application 10/821,832

11
CLAIM 5
Claim 5 recites the computer process further simplifies host groups
and LUN security groups into virtual connections for analysis. Appellants
contend that the Examiner fails to demonstrate how Nahum’s authentication
process for hosts disclosed in paragraph 18 and discussed above teaches
simplifying host groups and LUN security groups into virtual connections
for analysis. Ans. 10, 20-21. We agree. The Examiner fails to articulate
how Nahum’s security procedure for authenticating hosts teaches or suggests
simplifying groups into virtual connections for analysis.
For the foregoing reasons, Appellants have persuaded us of error in
the rejection of claim 5.
CLAIMS 8, 10, 15, 22, AND 25
For the rejection of the remaining claims 8, 10, 15, 22, and 25,
Appellants refer to the previous arguments of independent claims 6, 11, and
20 from which each remaining claim depends directly. See App. Br. 14-15.
We are not persuaded by Appellants’ arguments for the reasons discussed
above and need not address whether Nahum cures any purported deficiency.
For the foregoing reasons, Appellants have not persuaded us of error in the
rejections of claims 8, 10, 15, 22, and 25.

CONCLUSION
Under § 103, the Examiner did not err in rejecting claims 1, 2, 4, 6-17,
and 20-26 but did err in rejecting claims 3, 5, 18, and 19.

DECISION
The Examiner’s decision rejecting claims 1-26 is affirmed-in-part.
Appeal 2010-004979
Application 10/821,832

12
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

babc