Ex Parte Cardona et al

Patent Trial and Appeal BoardMay 19, 2017

14566253 (P.T.A.B. May. 19, 2017)

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. 14/566,253 12/10/2014 Omar Cardona AUS920100042US3 3401 50170 7590 05/19/2017 IBM CORP. (WIP) c/o WALDER INTELLECTUAL PROPERTY LAW, P.C. 17304 PRESTON ROAD SUITE 200 DALLAS, TX 75252 EXAMINER ONAT, UMUT ART UNIT PAPER NUMBER 2194 MAIL DATE DELIVERY MODE 05/19/2017 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 OMAR CARDONA, JAMES B. CUNNINGHAM, MATTHEW R. OCHS, and RAKESH SHARMA ____________ Appeal 2017-003522 Application 14/566,253 Technology Center 2100 ____________ Before MAHSHID D. SAADAT, ALLEN R. MacDONALD, and JOHN P. PINKERTON, Administrative Patent Judges. SAADAT, Administrative Patent Judge. DECISION ON APPEAL Appellants1 appeal under 35 U.S.C. § 134(a) from the Examiner’s Final Rejection of claims 1, 3–5, 8, 10–12, 15, and 17–19.2 We have jurisdiction under 35 U.S.C. § 6(b). We affirm. 1 According to Appellants, the real party in interest is International Business Machines Corporation (App. Br. 2). 2 Claims 7, 14, and 21 were objected to as being dependent upon a rejected base claim, but were also indicated as being allowable if rewritten in independent form including all of the limitations of the base claim (Final Act. 15). Appeal 2017-003522 Application 14/566,253 2 STATEMENT OF THE CASE Appellants’ invention relates to a mechanism that provides resource affinity for active memory sharing (AMS) and central processing unit (CPU) utilization via dynamic reconfiguration of the underlying hardware (Spec. ¶ 14). Exemplary claim 1 under appeal reads as follows: 1. A method, in a data processing system, for providing resource affinity for multi-queue network adapters via dynamic reconfiguration, the method comprising: allocating, by a device driver in the data processing system, an initial queue pair within a memory, wherein each queue pair is a transmit/receive queue pair; determining, by the device driver, whether workload of the data processing system has risen above a predetermined high threshold; responsive to the workload rising above the predetermined high threshold, allocating and initializing, by the device driver, an additional queue pair in the memory; programming, by the device driver, a receive side scaling (RSS) mechanism in a network adapter to allow for dynamic insertion of an additional processing engine associated with the additional queue pair; enabling, by the device driver, transmit tuple hashing to the additional queue pair; and repeating, by the device driver, the allocating and initializing, programming, and enabling steps each time the workload of the data processing system rises above the predetermined high threshold. REFERENCES and REJECTION Claims 1, 3–5, 8, 10–12, 15, and 17–19 stand rejected under 35 U.S.C. § 103(a) as unpatentable over Dubal et al. (US 2011/0142064 A1; Appeal 2017-003522 Application 14/566,253 3 published June 16, 2011) (“Dubal”) and Li (US 2011/0153935 A1; published June 23, 2011) (“Li”). ANALYSIS Independent Claims 1, 8, and 15 First Issue Appellants contend Dubal merely describes assigning a receive queue to a packet flow associated with a network application in a host system, and fails to teach or suggest allocating a transmit queue to form a transmit/receive queue pair as recited in the independent claims (see App. Br. 7–8; see also Reply Br. 2). Appellants further contend, although Li teaches transmit/receive queue pairs, Li describes that the transmit/receive queue pairs are allocated based on a speed of a network adapter and/or communication link, where the claimed invention changes the allocation of transmit/receive queue pairs based on workload (see App. Br. 9–10; see also Reply Br. 3–4). This argument is not persuasive, as the argument attacks Dubal and Li individually, rather than the combination of references. More specifically, Appellants’ argument that Dubal fails to teach a transmit queue (and thus, also fails to teach a transmit/receive queue pair) is not persuasive because the Examiner relied upon Li for teaching a transmit/receive queue pair (see Ans. 5). Similarly, Appellants’ further argument that Li fails to teach or suggest allocating an additional queue in response to a workload exceeding a threshold is unpersuasive because the Examiner relied upon Dubal for teaching allocating an additional queue in response to a workload exceeding a threshold (see id). It is well established that one cannot show non- Appeal 2017-003522 Application 14/566,253 4 obviousness by attacking references individually where the rejection is based upon the teachings of a combination of references. See In re Merck & Co., 800 F.2d 1091, 1097 (Fed. Cir. 1986); see also In re Keller 642 F.2d 413, 425 (CCPA 1981). In other words, Appellants’ argument is not persuasive because it does not address the actual reasoning of the Examiner’s rejection. Thus, we agree with the Examiner that the combination of Dubal and Li teaches allocating a transmit/receive queue pair as recited in the independent claims. Second Issue Appellants contend Dubal merely describes migrating an application to an existing queue, rather than a newly created queue, and, thus, Dubal’s migration of the application fails to teach or suggest allocating and initializing an additional queue pair in a memory as recited in the claims (see App. Br. 11–12). This argument is not persuasive. A claim under examination is given its broadest reasonable interpretation consistent with the underlying specification. See In re American Acad. of Science Tech. Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004). In the absence of an express definition of a claim term in the specification or a clear disclaimer of scope, the claim term is interpreted as broadly as the ordinary usage of the term by one of ordinary skill in the art would permit. See In re ICON Health & Fitness, Inc., 496 F.3d 1374, 1379 (Fed. Cir. 2007); see also In re Morris, 127 F.3d 1048, 1054 (Fed. Cir. 1997). We agree with the Examiner that Appellants’ Specification does not provide any description regarding the claim terms “allocating” and “initializing” that would limit the phrase “allocating and initializing” to mean “creating,” as argued by Appellants (see Ans. 14). The Examiner properly interprets “allocating and initializing Appeal 2017-003522 Application 14/566,253 5 . . . an additional [queue] in the memory” broadly but reasonably in light of Appellants’ Specification to encompass Dubal’s migration of an application from an original existing queue to a new existing queue. Further, as described above, Li teaches a queue pair (see Li ¶ 31). Thus, we agree with the Examiner that the combination of Dubal and Li teaches allocating and initializing an additional queue pair in a memory as recited in the independent claims. Third Issue Appellants contend Dubal merely describes migrating an application to an existing central processing unit (“CPU”), rather than a newly created CPU, and, thus, Dubal’s migration of the application fails to teach or suggest a dynamic insertion of an additional processing engine associated with an additional queue pair as recited in the claims (see App. Br. 13). This argument is not persuasive. The Examiner finds that Dubal teaches migrating an application from an existing core to a “new core” (i.e., from an original existing CPU to a new existing CPU) via a receive side scaling (“RSS”) mechanism (see Ans. 17–18 (citing Dubal ¶¶ 13, 23)). Appellants’ Specification does not provide any description regarding the claim term “additional processing engine” that would limit the aforementioned term to mean “new processing engine,” as argued by Appellants. Thus, the Examiner’s interpretation of “dynamic insertion of an additional processing engine” as reading on Dubal’s migration of an application from an original existing CPU to a new existing CPU, while broad, is reasonable in light of Appellants’ Specification. Further, as described above, Li teaches a queue pair (see Li ¶ 31). Thus, we agree with the Examiner that the combination of Dubal and Li teaches a dynamic insertion of an additional processing Appeal 2017-003522 Application 14/566,253 6 engine associated with the additional queue pair as recited in the independent claims. Fourth Issue Appellants contend Dubal merely describes hashing, and fails to teach or suggest enabling transmit tuple hashing to an additional queue pair as recited in the claims (see App. Br. 16). This argument is not persuasive in view of the findings and reasons provided by the Examiner in response to Appellants’ argument (see Ans. 18–19). Thus, we agree with the Examiner that the combination of Dubal and Li teaches enabling transmit tuple hashing to the additional queue pair as recited in the independent claims. Fifth Issue Appellants contend Dubal merely describes performing an assignment of an application dynamically during an operation of a system and/or as one or more applications become invoked in a memory, and Dubal’s repeating steps when one or more applications become invoked in the memory fails to teach or suggest repeating allocating and initializing, programming, and enabling each time a workload of a computing device rises above a predetermined high threshold as recited in the claims (see App. Br. 20). This argument is not persuasive in view of the findings and reasons provided by the Examiner in response to Appellants’ argument (see Ans. 19–20). Thus, we agree with the Examiner that the combination of Dubal and Li teaches repeating the allocating and initializing, programming, and enabling steps each time a workload of a data processing system rises above a predetermined threshold as recited in the independent claims. Appeal 2017-003522 Application 14/566,253 7 Accordingly, we sustain the Examiner’s rejection of independent claims 1, 8, and 15 under 35 U.S.C. § 103(a) as unpatentable over Dubal and Li. Remaining Claims No separate arguments are presented for the remaining dependent claims (see App. Br. 20). We therefore sustain their rejections for the reasons stated with respect to independent claims 1, 8, and 15. DECISION We affirm the Examiner’s decision to reject claims 1, 3–5, 8, 10–12, 15, and 17–19. 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