Prosecution Insights
Last updated: April 19, 2026
Application No. 16/786,602

COMPUTE GRAPH OPTIMIZATION

Non-Final OA §103
Filed
Feb 10, 2020
Examiner
KHONG, ALEXANDER
Art Unit
2168
Tech Center
2100 — Computer Architecture & Software
Assignee
Nvidia Corporation
OA Round
7 (Non-Final)
84%
Grant Probability
Favorable
7-8
OA Rounds
2y 7m
To Grant
99%
With Interview

Examiner Intelligence

Grants 84% — above average
84%
Career Allow Rate
543 granted / 646 resolved
+29.1% vs TC avg
Strong +28% interview lift
Without
With
+27.9%
Interview Lift
resolved cases with interview
Typical timeline
2y 7m
Avg Prosecution
15 currently pending
Career history
661
Total Applications
across all art units

Statute-Specific Performance

§101
11.0%
-29.0% vs TC avg
§103
49.7%
+9.7% vs TC avg
§102
17.1%
-22.9% vs TC avg
§112
9.3%
-30.7% vs TC avg
Black line = Tech Center average estimate • Based on career data from 646 resolved cases

Office Action

§103
Notice of Pre-AIA or AIA Status The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Continued Examination Under 37 CFR 1.114 A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12/12/2025 has been entered. Response to Arguments Applicant's arguments with respect to the claims have been considered but are moot in view of the new ground(s) of rejection. The independent claims have been amended to include new limitations that were never previously presented; thereby, these amendments change the scope of the claims. However, newly found prior arts are applied. Claim Objections Claim 2 is objected to because of the following informalities: in claim 2, the word “and” in the limitation “and OpenCL graph” appears to be a typo. Appropriate correction is required. Claim Rejections - 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claim(s) 1, 3-4, 7, 17, 20, 22, 25-27 and 30-32 is/are rejected under 35 U.S.C. 103 as being unpatentable over Capalija et al. (U.S. PG Pub. No. 20200401402 A1, hereinafter “Capalija”) in view of Vercellotti et al. (U.S. PG Pub. 20210200505 A1, hereinafter “Vercellotti”). Regarding claim 1, Capalija teaches a processor, comprising: one or more circuits to: identify one or more portions of an of processing core operational codes and a set of operand identifiers to execute the directed graph and a set of operand identifiers”); perform the one or more optimizations on the one or more identified portions (Capalija ¶0041, “The compiler can also output an assignment of packets for storage on specific processing cores within the set of processing cores to initialize the directed graph for execution and breakdown the instructions of the application code into instructions for execution on individual processing cores in such a way that the execution of the application code is maximally parallelized with minimal memory latency and such that data movement is optimized for that purpose”); and cause the Capalija fails to explicitly teaches the graph is an instantiated graph and the identification based on part of the changes made subsequent to instantiation. However, in the same field of endeavor, Vercellotti teaches the graph is an instantiated graph and the identification based on part of the changes made subsequent to instantiation (Vercellotti ¶0030, “The controller then instantiates the optimized signal graph in the system hardware, software, and available connections (FIG. 1, 116). This may be performed iteratively or as the optimized graph is developed, or as a separate step after the optimization process has been completed. Graph recomposition (106) and instantiation in hardware/software (116) may be merged into a single step in a system which instantiates all graph changes directly while the graph is being optimized”, “The controller maintains internal representation 232 that is updated to reflect the user's current requirements. Having determined that an audio signal graph recomposition is required, the controller performs graph configuration step 106 (FIG. 1) to generate an optimized graph that meets the user's requirements subject to the system constraints”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Capalija by incorporating the teachings of Vercellotti. The motivation would be for providing an optimized graph that meet user’s requirements subject to the system constraints (Vercellotti ¶0031). As to claim 3, Capalija as modified by Vercellotti also teaches the processor of claim 1, wherein the one or more optimizations for executing the instantiated graph are reused the instantiated graph is performed different times using the one or more changes to the one or more operands (Capalija ¶0041, “The compiler can also output an assignment of packets for storage on specific processing cores within the set of processing cores to initialize the directed graph for execution and breakdown the instructions of the application code into instructions for execution on individual processing cores in such a way that the execution of the application code is maximally parallelized with minimal memory latency and such that data movement is optimized for that purpose”). As to claim 4, Capalija as modified by Vercellotti also teaches also teaches the processor of claim 1, wherein the instantiated graph is optimized, prior to a first performance of the instantiated graph, for execution on a processing unit, and the optimized instantiated graph is reused (Capalija ¶0041, i.e., optimized instantiated graph is reused after initializing). As to claim 7, Capalija as modified by Vercellotti also teaches the processor of claim 1, the one or more circuits to determine that the instantiated graph can be performed, using the one or more changes to the one or more operands, without changing a topology for executing the instantiated graph (Capalija ¶0042, i.e., optimized instantiated graph is performed after initializing). Claim 17 recites the limitations substantially similar to those of claim 1 and is similarly rejected. Claim 20 recites the limitations substantially similar to those of claim 7 and is similarly rejected. As to claim 22, Capalija as modified by Vercellotti also teaches the system of claim 17, the one or more processors to: generate optimized instructions for performing the instantiated graph a first time based on a first set of the one or more operands (Capalija ¶0041); and perform the instantiated graph a second time, using a second set of the one or more operands, using the optimized instructions (Capalija ¶¶0041-0042). Regarding claim 25, Capalija as modified by Vercellotti also teaches a system comprising: a memory comprising a first buffer and a second buffer (Capalija ¶0054); and at least one processor to perform the same method as recited in claim 1. Claim 25 is similarly rejected. As to claim 26, Capalija as modified by Vercellotti also teaches the system of claim 25, wherein the instantiated graph is optimized based at least in part on the first buffer (Capalija ¶0042 and ¶0054). As to claim 27, Capalija as modified by Vercellotti also teaches the system of claim 26, wherein optimizations of the instantiated graph based at least in part on the first buffer operates are used to perform the instantiated graph the second time (Capalija ¶0042 and ¶0054). As to claim 30, Capalija as modified by Vercellotti also teaches the system of claim 25, wherein the instantiated graph is optimized upon being instantiated prior to a first time the instantiated graph is performed (Capalija ¶0041). As to claim 31, Capalija as modified by Vercellotti also teaches the system of claim 25 but fails to explicitly teach wherein the instantiated graph is optimized for performance on selected processing unit (Capalija ¶0041). As to claim 32, Capalija as modified by Vercellotti also teaches the processor of claim 1, wherein the instantiated graph is compiled prior to the multiple executions and at least part of the instantiated graph is reused (Capalija ¶0041). Claims 2, 18 and 28-29 is/are rejected under 35 U.S.C. 103 as being unpatentable over Capalija and Vercellotti, and further in view of Liu et al. (U.S. PG Pub. No. 2020/0334083 A1, hereinafter “Liu”). As to claim 2, Capalija as modified by Vercellotti teaches the processor of claim 1, but failed to explicitly teach wherein the instantiated graph is at least one of a compute unified architecture, (“CUDA”) graph, and OpenCL graph, or a heterogeneous compute interface for portability (“HIP”) graph. However, in the same field of endeavor, Liu teaches the instantiated graph is at least one of a compute unified architecture (“CUDA”) graph (Liu ¶0035) and OpenCL graph, or a heterogeneous compute interface for portability (“HIP”) graph. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Capalija and Vercellotti by incorporating the teachings of Liu. The motivation would be for tuning CUDA application performance (Liu ¶0036). Claim 18 recites the limitations substantially similar to those of claim 2 and is similarly rejected. As to claim 28, Capalija as modified by Vercellotti also teaches the system of claim 25, but fails to explicitly teach wherein the instantiated graph is a CUDA graph. However, in the same field of endeavor, Liu teaches the instantiated graph is a CUDA graph (Liu ¶0035). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Capalija and Vercellotti by incorporating the teachings of Liu. The motivation would be for tuning CUDA application performance (Liu ¶0036). As to claim 29, Capalija as modified by Vercellotti and Liu also teaches the system of claim 28, wherein the instantiated graph comprises a plurality of CUDA kernels (Liu ¶0035). Claims 6 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Capalija and Vercellotti, and further in view of Fleming et al. (U.S. Patent No. 10,416,999 B2, hereinafter “Fleming”). As to claim 6, Capalija as modified by Vercellotti teaches the processor of claim 1 but fails to explicitly teach wherein the one or more operands comprise a first buffer operated on by the instantiated graph when the instantiated graph is performed a first time, and a second buffer, different than the first buffer, operated on by the instantiated graph when the instantiated graph is performed a second time. However, in the same field of endeavor, Fleming teaches the one or more operands comprise a first buffer operated on by the instantiated graph when the instantiated graph is performed a first time, and a second buffer, different than the first buffer, operated on by the instantiated graph when the instantiated graph is performed a second time (Fleming Claim 1, i.e., different input buffers are used for different data paths). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Capalija and Vercellotti by incorporating the teachings of Fleming. The motivation would be to provide a configurable spatial accelerator (Fleming Col 1 Ln 17-18). Claim 19 recites the limitations substantially similar to those of claim 6 and is similarly rejected. Allowable Subject Matter Claims 9-16 are allowed. Claims 22-24 are objected to as being dependent upon a rejected base claim(s), but would be allowable if rewritten in independent form including all of the limitations of the base claim(s) and any intervening claims. The features of claim limitations presented in claim 9, and in claims 8, 10-16 and 22-24 in combination with the other limitations recited in the context of their respective base claim(s) is allowable subject matter. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALEXANDER KHONG whose telephone number is (571)270-7127. The examiner can normally be reached Mon-Fri 8am-5pm EST. Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Charles Rones can be reached at (571)272-4085. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /ALEXANDER KHONG/Primary Examiner, Art Unit 2168
Read full office action

Prosecution Timeline

Feb 10, 2020
Application Filed
Dec 16, 2021
Non-Final Rejection — §103
Jun 06, 2022
Interview Requested
Jun 08, 2022
Applicant Interview (Telephonic)
Jun 08, 2022
Examiner Interview Summary
Jun 13, 2022
Response Filed
Jul 01, 2022
Final Rejection — §103
Nov 17, 2022
Interview Requested
Nov 29, 2022
Examiner Interview Summary
Nov 29, 2022
Applicant Interview (Telephonic)
Jan 06, 2023
Notice of Allowance
Mar 06, 2023
Response after Non-Final Action
Mar 28, 2023
Response after Non-Final Action
Jun 07, 2023
Non-Final Rejection — §103
Sep 13, 2023
Interview Requested
Sep 19, 2023
Applicant Interview (Telephonic)
Sep 19, 2023
Examiner Interview Summary
Dec 13, 2023
Response Filed
Feb 29, 2024
Final Rejection — §103
Sep 06, 2024
Notice of Allowance
Apr 07, 2025
Request for Continued Examination
Apr 09, 2025
Response after Non-Final Action
Apr 15, 2025
Non-Final Rejection — §103
Jun 04, 2025
Interview Requested
Jun 18, 2025
Applicant Interview (Telephonic)
Jun 18, 2025
Examiner Interview Summary
Jul 18, 2025
Response Filed
Aug 08, 2025
Final Rejection — §103
Dec 12, 2025
Request for Continued Examination
Dec 20, 2025
Response after Non-Final Action
Jan 07, 2026
Non-Final Rejection — §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12591592
METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR REPLICATING DATA
2y 5m to grant Granted Mar 31, 2026
Patent 12585546
DATA LINEAGE BASED MULTI-DATA STORE RECOVERY
2y 5m to grant Granted Mar 24, 2026
Patent 12579170
AUTOMATIC ORGANIZATION OF USER ACTIVITY INTO COLLECTIONS BASED ON TOPICS
2y 5m to grant Granted Mar 17, 2026
Patent 12579165
Restricted Blockchain Cluster
2y 5m to grant Granted Mar 17, 2026
Patent 12579518
SYSTEMS AND METHODS FOR PROVIDING CROSS-SECTIONAL SCALING
2y 5m to grant Granted Mar 17, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

AI Strategy Recommendation

Get an AI-powered prosecution strategy using examiner precedents, rejection analysis, and claim mapping.
Powered by AI — typically takes 5-10 seconds

Prosecution Projections

7-8
Expected OA Rounds
84%
Grant Probability
99%
With Interview (+27.9%)
2y 7m
Median Time to Grant
High
PTA Risk
Based on 646 resolved cases by this examiner. Grant probability derived from career allow rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month