International Business Machines Corporation

Patent Trials and Appeals Board

Feb 10, 2022

2020006496 (P.T.A.B. Feb. 10, 2022)

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. 14/966,561 12/11/2015 ALEXANDER E. MERICAS ROC920150345US1 6254 34533 7590 02/10/2022 IBM (AUS-KLS) c/o Kennedy Lenart Spraggins LLP 797 Sam Bass Road #2559 ROUND ROCK, TX 78681 EXAMINER KABIR, MOHAMMAD H ART UNIT PAPER NUMBER 2199 NOTIFICATION DATE DELIVERY MODE 02/10/2022 ELECTRONIC 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. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): eofficeaction@appcoll.com kate@klspatents.com office@klspatents.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE _______________ BEFORE THE PATENT TRIAL AND APPEAL BOARD _______________ Ex parte ALEXANDER E. MERICAS, MARIA L. PESANTEZ, and MYSORE S. SRINIVAS _______________ Appeal 2020-006496 Application 14/966,561 Technology Center 2100 _______________ Before JAMES R. HUGHES, JASON J. CHUNG, and JAMES W. DEJMEK, Administrative Patent Judges. DEJMEK, Administrative Patent Judge. DECISION ON APPEAL Appellant1 appeals under 35 U.S.C. § 134(a) from a Final Rejection of claims 8-20. Appellant has canceled claims 1-7. See Appeal Br. 16. We have jurisdiction over the remaining pending claims under 35 U.S.C. § 6(b). We reverse. 1 Throughout this Decision, we use the word “Appellant” to refer to “applicant” as defined in 37 C.F.R. § 1.42 (2019). Appellant identifies International Business Machines Corporation as the real party in interest. Appeal Br. 2. Appeal 2020-006496 Application 14/966,561 2 STATEMENT OF THE CASE Introduction Appellant’s disclosed and claimed invention generally relates to “instruction weighting for performance profiling in a group dispatch processor.” Spec. ¶ 4. According to the Specification, a group dispatch processor “dispatches and completes instructions according to a group.” Spec. ¶¶ 15, 26. In some embodiments, a group dispatch processor may also include a performance monitoring unit (“PMU”). Spec. ¶ 22, Fig. 1. The PMU is configured to provide “detailed information descriptive of the utilization of instruction execution resources and storage control.” Spec. ¶ 27. For example, the PMU may analyze a sample of executed code to identify or monitor performance bottlenecks, idle cycles, stalls, and overall efficiency. See Spec. ¶ 27. According to the Specification, the captured performance information may be used to optimize the architecture of the processor and the instructions being executed. Spec. ¶ 27. Claim 8 is illustrative of the subject matter on appeal and is reproduced below with the disputed limitations emphasized in italics: 8. An apparatus for instruction weighting for performance profiling in a group dispatch processor that dispatches and completes instructions according to dispatch groups, the apparatus comprising a computer processor and a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the steps of: retrieving, by a post processing profiler, an execution sample of completed instructions, wherein the execution sample includes: an instruction address of a youngest instruction in a dispatch group that has completed execution in a group Appeal 2020-006496 Application 14/966,561 3 dispatch processor, wherein the dispatch group is a group of instructions that are tagged by the group dispatch processor, dispatched by the group dispatch processor, and completed according to the dispatch group; a number of instructions in the dispatch group; and a number of times that the instruction address of the youngest instruction in the dispatch group address is captured; based on the instruction address of the youngest instruction and the number of instructions in the dispatch group, identifying, by the post processing profiler, all of the instructions that are in the dispatch group at a time that the dispatch group completed execution; and weighting, by the post processing profiler, all of the identified instructions in the dispatch group by applying within an execution profile the result of the execution sample to all of the identified instructions that are in the dispatch group and that were completed according to the dispatch group by calculating a percentage of the result of the execution sample to be attributed to the instructions in the dispatch group and assigning to each of the identified instructions that are in the dispatch group an equal portion of the percentage. The Examiner’s Rejections2 1. Claims 8, 9, 12, 14-16, 19, and 20 stand rejected under 35 U.S.C. § 103 as being unpatentable over DeSylva et al. (US 2005/0071613 A1; Mar. 31, 2005) (“DeSylva”); Blumenthal et al. (US 7,546,598 B2; June 9, 2009) (“Blumenthal”); Anderson et al. (US 6,092,180; 2 The Examiner had rejected claims 8-14 under 35 U.S.C. § 112(b) as being indefinite. See Final Act. 7. The Examiner has withdrawn this rejection. See Ans. 3. Appeal 2020-006496 Application 14/966,561 4 July 18, 2000) (“Anderson”); and Henzinger et al. (US 5,857,097; Jan. 5, 1999) (“Henzinger”). Final Act. 9-18. 2. Claims 10 and 17 stand rejected under 35 U.S.C. § 103 as being unpatentable over DeSylva, Blumenthal, Anderson, Henzinger, and DeWitt, JR et al. (US 2005/0154812 A1; July 14, 2005) (“DeWitt”). Final Act. 19- 20. 3. Claims 11 and 18 stand rejected under 35 U.S.C. § 103 as being unpatentable over DeSylva, Blumenthal, Anderson, Henzinger, DeWitt, and Kosche et al. (US 2008/0177756 A1; July 24, 2008) (“Kosche”). Final Act. 20-22. 4. Claim 13 stands rejected under 35 U.S.C. § 103 as being unpatentable over DeSylva, Blumenthal, Anderson, Henzinger, and Doing et al. (US 8,479,184 B2; July 2, 2013) (“Doing”). Final Act. 22-23. ANALYSIS3 In rejecting independent claim 8 (as well as independent claim 15), the Examiner relies on the combined teachings of DeSylva, Blumenthal, Anderson, and Henzinger. See Final Act. 9-16, 18. In relevant part, the Examiner finds DeSylva teaches retrieving an execution sample of completed instructions, wherein the execution sample includes an instruction address of a youngest instruction in a dispatch group that has completed 3 Throughout this Decision, we have considered the Appeal Brief, filed January 21, 2020 (“Appeal Br.”); the Examiner’s Answer, mailed April 29, 2020 (“Ans.”); and the Final Office Action, mailed October 21, 2019 (“Final Act.”), from which this Appeal is taken. Appellant did not file a Reply Brief. To the extent Appellant has not advanced separate, substantive arguments for particular claims or issues, such arguments are considered waived. See 37 C.F.R. § 41.37(c)(1)(iv). Appeal 2020-006496 Application 14/966,561 5 execution and a number of instructions in the dispatch group. See Final Act. 9-11 (citing DeSylva ¶¶ 24, 29-30). Further, the Examiner finds DeSylva teaches identifying, by a post processing profiler, all of the instructions in the dispatch group that completed execution. See Final Act. 11-12 (citing DeSylva ¶¶ 13, 29, 33, Figs. 1, 4). The Examiner also finds Blumenthal teaches weighting all of the identified instructions in the dispatch group by applying within an exception profile the result of the execution sample to all of the identified instructions in the dispatch group that were completed. See Final Act. 12-13 (citing Blumenthal, col. 3, ll. 1- 11, col. 7, ll. 18-24). In addition, the Examiner relies on Anderson to teach that the application of the result of the execution sample is performed by calculating a percentage of the result of the execution sample to be attributed to the instructions in the dispatch group and assigning to each of the identified instructions that are in the dispatch group an equal portion of the percentage. See Final Act. 13-14 (citing Anderson, col. 21, ll. 34-51, col. 22, ll. 10-15). Appellant asserts, inter alia, that Anderson, as relied on by the Examiner, fails to teach calculating a percentage of the execution sample result to be attributed to the identified instructions in the dispatch group, as recited in claim 8. Appeal Br. 11-13. More specifically, Appellant argues that “Anderson make no mention of calculating a percentage of the execution sample result that is to be attributed to the identified instructions in the dispatch group.” Appeal Br. 12. More specifically, Appellant asserts that Anderson describes a percentage of wasted issue slots, but makes no mention of a percentage of an execution sample result. Appeal Br. 12. Appeal 2020-006496 Application 14/966,561 6 In response, the Examiner finds Anderson discloses a “sample count and execution.” Ans. 7-8 (citing Anderson, Fig. 5, Abstract). More specifically, the Examiner finds that Anderson teaches profile information may be analyzed to measure concurrency levels. Ans. 8 (citing Anderson, col. 9, ll. 50-63). Further, the Examiner finds Anderson describes that “it may be useful to compute the average concurrency level when instruction I ‘hits’ in one of the caches, and then to compare the average concurrency level with the case where instruction I suffers a cache miss.” Ans. 8 (quoting Anderson, col. 22, ll. 39-48). From these teachings, the Examiner finds Anderson teaches calculating a percentage of the result of the exception sample to be attributed to the instructions in the dispatch group and assigning to each of the identified instructions an equal portion of the percentage. See Ans. 8-9. Contrary to the Examiner’s findings, we do not find the cited portions of Anderson teach or reasonably suggest calculating a percentage of the result of the execution sample to be attributed to instructions in the dispatch group and further assigning an equal portion of the percentage to the identified instructions. Although Anderson generally relates to “randomly selecting instructions, and sampling detailed state information for the selected instructions,” the Examiner has not provided sufficient evidence or technical reasoning to show that Anderson further teaches calculating a percentage of the result of the execution sample (i.e., the selected instruction) attributed to the instruction in a dispatch group and assigning an equal portion of the percentage to each of the identified instructions in the dispatch group. As identified by Appellant, Anderson’s discussion regarding percentages relates to the percentage of wasted slots per execution Appeal 2020-006496 Application 14/966,561 7 of instruction I. See Anderson, col. 21, ll. 34-51. In addition, although Anderson’s computation of an average concurrency level when an instruction hits one of the caches as compared to the average concurrency level when an instruction misses the cache could perhaps be expressed as a percentage, we do not see how the cited portion of Anderson (alone or in combination with the other references) applies an equal portion of such a percentage to all of the identified instructions in a dispatch group. Because we find it dispositive that Anderson, as relied on by the Examiner, does not teach calculating a percentage of the result of the execution sample to be attributed to instructions in the dispatch group and further assigning an equal portion of the percentage to the identified instructions in a dispatch group as claimed, we do not address other issues raised by Appellant’s arguments. See Beloit Corp. v. Valmet Oy, 742 F.2d 1421, 1423 (Fed. Cir. 1984) (finding an administrative agency is at liberty to reach a decision based on “a single dispositive issue”). For the reasons discussed supra, we are persuaded of Examiner error. Accordingly, we do not sustain the Examiner’s rejection of independent claim 8. For similar reasons, we do not sustain the Examiner’s rejection of independent claim 15, which recites commensurate limitations. Additionally, we do not sustain the Examiner’s rejection of claims 8-14 and 16-20, which depend directly or indirectly therefrom. CONCLUSION We reverse the Examiner’s decision rejecting claims 8-20 under 35 U.S.C. § 103. Appeal 2020-006496 Application 14/966,561 8 DECISION SUMMARY Claim(s) Rejected 35 U.S.C. § Reference(s)/Basis Affirmed Reversed 8, 9, 12, 14-16, 19, 20 103 DeSylva, Blumenthal, Anderson, Henzinger 8, 9, 12, 14-16, 19, 20 10, 17 103 DeSylva, Blumenthal, Anderson, Henzinger, DeWitt 10, 17 11, 18 103 DeSylva, Blumenthal, Anderson, Henzinger, DeWitt, Kosche 11, 18 13 103 DeSylva, Blumenthal, Anderson, Henzinger, Doing 13 Overall Outcome 8-20 REVERSED