DETAILED ACTION
Claims 1-18 are pending.
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 .
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1, 2, 6-8, 12-14 and 18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Kanoun et al. (Online Energy-Efficient Task-Graph Scheduling for Multicore Platforms – cited in the IDS filed 5/19/2026).
As to claim 1, Kanoun teaches a multi-core processor, comprising multiple processing cores (A multi-core platform; see Fig. 2 and M cores; page 4, right column, 1st paragraph), wherein the multi-core processor executes programming instructions to:
store multiple job chains and dependency relationships between the multiple job chains, wherein the dependency relationship comprises dependency and non-dependency (see figure 1,"Connected DAGs" are dependent job chains as claimed and "independent DAGs" are independent job chains as claimed, see also page 1, right column, first paragraph: "DAGs where each node denotes a task");
determine a first job chain and a second job chain in the multiple job chains based on the dependency relationships between the multiple job chains, wherein a dependency relationship between the first job chain and the second job chain is non-dependency, the first job chain comprises one or more first tasks, and the second job chain comprises one or more second tasks (figure 1, "independent DAGs", page 2, right column: "2) It does not impose any restrictions on the DAG (see restrictions on deadline dependencies as in the fork-join model). Our scheduler covers online all DAG models (Fig. 1).");
schedule a part or all of the multiple processing cores to execute the one or more first tasks (figure 2 and page 2, right column: "Our scheduler covers on line all DAG models (Fig. 1 ")); and
while executing the one or more first tasks on the part or all of the multiple processing cores, schedule at least one second task in the second job chain to at least one first processing core for execution when the at least one first processing core in the multiple processing cores is in an idle state (page 4, last 5 lines: "gaps (i.e., when a core is idle and waiting for another task to finish)", figure 5(c), "Filling the gaps" and page 5, under section C: "mapping tasks from future deadlines (i.e., Te+I with O < I < N) to the gaps generated by the first scheduler."), and page 7, section F, “same depth level le,0 because there are no dependencies between tasks at the same depth. Therefore, If a core becomes available and there are no tasks ready to be scheduled from Te, a gap is then detected on the available core as all the remaining tasks of Te start from le,k with k ≥ 1 and depend on at least one of the tasks currently running in le,0.”).
As to claim 2, Kanoun teaches determining that the dependency relationship between the first job chain and the second job chain is non-dependency (page 4, first paragraph: "We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis.", in other words, the algorithm stores the dependency relationship between DAGs in the form deadline dependencies, which discloses a dependency management unit as claimed, and the WS buffer stores task sets with different deadlines).
As to claim 6, Kanoun teaches obtain a command stream and dependency relationships between a part or all of the multiple job chains (see page 3, Section A: “We model computationally intensive applications (see [13] and [17]) as a DAG G =< N, E > of dependent tasks tj with nondeterministic workload wj and coarse-grained soft deadlines. N is the node set containing all the tasks. E is the edge set, which models the dependencies among the tasks. Each node in the DAG denotes a task tj. Ekj denotes that there is a directed edge from tj to tk indicating that task k depends on task j. Each task tj is characterized with its index j and a deadline di. Our solution allows coarse-grained deadlines where a deadline can be assigned to a subset of tasks indexed by i. Our model covers all general DAG models (see all DAG models of Fig. 1) including the general case where a task’s children may have different deadlines than the task itself and its other children (i.e., DAG model 4 of Fig. 1)”); and generate the part or all of the multiple job chains based on the command stream (see Page 4, left column, 1st paragraph: “We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis. The DFM processes the full DAG of the application using this WS buffer where only a limited number of deadlines are monitored at a time.”).
As to claim 7, see rejection of claim 1 above.
Kanoun further teaches a method (solution; abstract), applied to a multi-core processor (A multi-core platform; see Fig. 2 and M cores; page 4, right column, 1st paragraph), wherein the multi-core processor comprises a job manager (DAG flow manager; see Fig. 2) and multiple processing cores coupled to the job manager (see Fig. 2).
As to claim 8, Kanoun teaches the method according to claim 7, wherein the job manager comprises a dependency manager (DAG flow manager; see Fig. 2 – DAG flow manager includes functions of the dependency manager) and a task queue (buffer; see Fig. 2);
wherein storing the dependency relationships between the multiple job chains by using the job manager comprises:
storing the dependency relationships between the multiple job chains by using the dependency manager in the job manager (page 4, first paragraph: "We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis.", in other words, the algorithm stores the dependency relationship between DAGs in the form deadline dependencies, which discloses a dependency management unit as claimed, and the WS buffer stores task sets with different deadlines); and
wherein determining the first job chain and the second job chain in the multiple job chains based on the dependency relationships between the multiple job chains by using the job manager comprises:
sending a first instruction to the task queue by using the dependency manager in the job manager if determining, by using the dependency manager in the job manager, that the dependency relationship between the first job chain and the second job chain is non-dependency, wherein the first instruction indicates that the dependency relationship between the first job chain and the second job chain is non-dependency (page 4, first paragraph: "We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis.", in other words, the algorithm stores the dependency relationship between DAGs in the form deadline dependencies, which discloses a dependency management unit as claimed, and the WS buffer stores task sets with different deadlines, i.e. it is a task queue unit receiving a first instruction from the dependency management unit as claimed).
As to claim 12, Kanoun teaches the method according to claim 8, wherein the job manager further comprises a task assembling manager (DAG Flow Manager … In our solution, we integrate the DFM that we proposed in [21] to monitor general DAG models and to prepare a set of outputs for our scheduler; page 3, section C), and the method further comprises:
obtaining a command stream and dependency relationships between a part or all of the multiple job chains by using the task assembling manager in the job manager, and generating the part or all of the multiple job chains based on the command stream (We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis. Each Ti is associated with an adjacency matrix, a deadline and a list of edges connecting it with Ti+l (with l _= 0). This information must be provided to the DFM by the application. Analyzing the full DAG of an application by subsets of N deadlines (see Tis) is the key to having a low complexity online DAG monitoring solution; page 4, left column, 1st paragraph); and
sending the part or all of the multiple job chains to the task queue by using the task assembling manager in the job manager, and sending the dependency relationships between the part or all of the multiple job chains to the dependency manager (The DFM processes the full DAG of the application using this WS buffer where only a limited number of deadlines are monitored at a time. When all of the tasks in deadline task set Ti finish executing, the DFM requests the next Ti input from the application; page 4, left column, 1st paragraph).
As to claim 13, see rejection of claim 1 above.
Kanoun further teaches a multi-core processor (A multi-core platform; see Fig. 2 and M cores; page 4, right column, 1st paragraph), wherein the multi-core processor comprises a job manager (DAG flow manager; see Fig. 2) and multiple processing cores coupled to the job manager (see Fig. 2).
As to claims 14 and 18, see rejections of claims 8 and 12 above, respectively.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 3-5, 9-11 and 15-17 are rejected under 35 U.S.C. 103 as being unpatentable over Kanoun et al. (Online Energy-Efficient Task-Graph Scheduling for Multicore Platforms – cited in the IDS filed 5/19/2026) in view of VRIND et al. (US 2019/0087224 A1).
As to claim 3, Kanoun does not teach preempt a processing core for the first job chain and the second job chain.
However, VRIND, in the same field of endeavor, teaches a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core (paragraphs [0032] and [0035]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teaching of VRIND to the system of Kanoun because VRIND teaches a method that tasks in a system can be executed based on priority-driven scheduling (paragraph [0032]).
As to claim 4, Kanoun teaches split the first job chain into the one or more first tasks (see Kanoun: Each job Ji can be further partitioned into ni sub-jobs running on several cores in parallel with the same deadline di; page 9, left column, 1st paragraph).
Kanoun does not teach preempt one or more second processing cores from the multiple processing cores; and scheduling the one or more second processing cores to execute the one or more first tasks.
However, VRIND, in the same field of endeavor, teaches a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core (paragraphs [0032] and [0035]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teaching of VRIND to the system of Kanoun, because VRIND teaches a method that tasks in a system can be executed based on priority-driven scheduling (paragraph [0032]). Thus, when applying the teaching of VRIND to the system of Kanoun, when the when one or more first tasks have higher priority than other tasks of the other stages, processors execute other lower tasks must be pre-emptive so the first tasks can be executed first.
As to claim 5, Kanoun teaches split the second job chain into the one or more second tasks (see Kanoun: Each job Ji can be further partitioned into ni sub-jobs running on several cores in parallel with the same deadline di; page 9, left column, 1st paragraph); and
scheduling the at least one second task in the one or more second tasks to the at least one first processing core for execution (see Kanoun: page 4, last 5 lines: "gaps (i.e., when a core is idle and waiting for another task to finish)", figure 5(c), "Filling the gaps" and page 5, under section C: "mapping tasks from future deadlines (i.e., Te+I with O < I < N) to the gaps generated by the first scheduler.").
Kanoun does not teach preempt the at least one first processing core when the at least one first processing core in the multiple processing cores is in the idle state.
However, VRIND, in the same field of endeavor, teaches a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core (paragraphs [0032] and [0035]).
As to claim 9, Kanoun as modified by VRIND teaches the method according to claim 8, wherein the job manager further comprises a task splitting manager (see Kanoun: Each job Ji can be further partitioned into ni sub-jobs running on several cores in parallel with the same deadline di; page 9, left column, 1st paragraph. Thus, the DAG Flow Manager includes logic for task splitting) and a multi-core manager (see Kanoun: multi-core; see Fig. 2);
wherein storing the multiple job chains by using the job manager comprises:
storing the multiple job chains by using the task queue in the job manager (see Kanoun: page 4, first paragraph: "We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis, the WS buffer stores task sets with different deadlines, i.e. it is a task queue unit receiving a first instruction from the dependency management unit as claimed); and
wherein determining the first job chain and the second job chain in the multiple job chains based on the dependency relationships between the multiple job chains by using the job manager further comprises:
after receiving, by using the task queue in the job manager, the first instruction sent by
using the dependency manager in the job manager, sending the first job chain and the second job
chain to the task splitting manager by using the task queue in the job manager (see Kanoun: page 4, first paragraph: "We define Ti as the subset of tasks having the same deadline di (we refer to Ti as a deadline task set). We also define the Working Set WS as a look-ahead window buffer with N Tis.", in other words, the algorithm stores the dependency relationship between DAGs in the form deadline dependencies, which discloses a dependency management unit as claimed, and the WS buffer stores task sets with different deadlines, i.e. it is a task queue unit receiving a first instruction from the dependency management unit as claimed); and
sending a second instruction to the multi-core manager, wherein the second instruction indicates the multi-core manager to preempt a processing core for the first job chain and the second job chain (see VRIND: a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core; paragraphs [0032] and [0035]).
As to claim 10, Kanoun teaches the method according to claim 9, wherein scheduling, by using the job manager, the part or all of the multiple processing cores to execute the one or more first tasks comprises:
splitting the first job chain into the one or more first tasks by using the task splitting
manager in the job manager (see Kanoun: Each job Ji can be further partitioned into ni sub-jobs running on several cores in parallel with the same deadline di; page 9, left column, 1st paragraph. Thus, the DAG Flow Manager includes logic for task splitting).
Kanoun does not teach preempting one or more second processing cores from the multiple processing cores based on the second instruction by using the multi-core manager in the job manager; sending, to the task splitting manager by using the multi-core manager in the job manager, a result of preempting the one or more second processing cores; and scheduling, by using the task splitting manager in the job manager, the one or more second processing cores to execute the one or more first tasks.
However, VRIND, in the same field of endeavor, teaches a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core (paragraphs [0032] and [0035]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teaching of VRIND to the system of Kanoun because VRIND teaches a method that tasks in a system can be executed based on priority-driven scheduling (paragraph [0032]). Thus, when applying the teaching of VRIND to the system of Kanoun, when the one or more first tasks have higher priority than other tasks of the other stages, processors execute other lower tasks must be pre-emptive so the first tasks can be executed first.
As to claim 11, Kanoun teaches the method according to claim 10, wherein scheduling, by using the job manager, the at least one second task in the second job chain to the at least one first processing core for execution when the at least one first processing core in the multiple processing cores is in the idle state comprises:
splitting the second job chain into the one or more second tasks by using the task splitting
manager in the job manager (see Kanoun: Each job Ji can be further partitioned into ni sub-jobs running on several cores in parallel with the same deadline di; page 9, left column, 1st paragraph. Thus, the DAG Flow Manager includes logic for task splitting).
Kanoun does not teach preempting the at least one first processing core based on the second instruction by using the multi-core manager in the job manager when the at least one first processing core in the multiple processing cores is in the idle state; sending, to the task splitting manager by using the multi-core manager in the job manager, a result of preempting the at least one first processing core; and scheduling, by using the task splitting manager in the job manager, the at least one second task in the one or more second tasks to the at least one first processing core for execution.
VRIND, in the same field of endeavor, teaches a pre-emptive task scheduling is adopted where highest-priority tasks are executed on any available processor core (paragraphs [0032] and [0035]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teaching of VRIND to the system of Kanoun because VRIND teaches a method that tasks in a system can be executed based on priority-driven scheduling (paragraph [0032]). Thus, when applying the teaching of VRIND to the system of Kanoun, when the one or more first tasks have higher priority than other tasks of the other stages, processors execute other lower tasks must be pre-emptive so the first tasks can be executed first.
As to claims 15-17, see rejections of claims 9-11 above, respectively.
Response to Arguments
Applicant’s arguments with respect to claims 1-18 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
Applicant's submission of an information disclosure statement under 37 CFR 1.97(c) with the timing fee set forth in 37 CFR 1.17(p) on 5/19/2026 prompted the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 609.04(b). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DIEM K CAO whose telephone number is (571)272-3760. The examiner can normally be reached Monday-Friday 8:00am-4:00pm.
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, April Blair can be reached at 571-270-1014. 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.
/DIEM K CAO/Primary Examiner, Art Unit 2196
DC
August 25, 2026
Prosecution Insights