DETAILED ACTION
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 .
Claims 1-4, 6-14, and 16-20 are pending. Claims 5 and 15 are cancelled.
Response to Arguments
Regarding: Prior Art Rejections:
Applicant’s amendments and arguments regarding the rejection of claims 1-4, 6-14, and 16-20 under 35 U.S.C. 103 have been fully considered and are moot due to new grounds of rejection necessitated by amendment.
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.
Claims 1-4, 7-14, and 17-20 are rejected under 35 U.S.C. 103 as being unpatentable over Richardson et al. US 11270220 B1 (“Richardson”) in view of Im et al. US 20110161965 A1 (“Im”) in view of Ravi et al. US 20240378085 A1 in view of Lee et al. US 20210357278 A1.
Richardson and Im are cited in a previous office action.
Regarding claim 1, Richardson teaches the invention substantially as claimed including:
A method, comprising:
quantum processing units in a telemetry plane (Col 6 lines 38-40 Quantum computing resources 160 may include computing devices that rely on quantum-mechanical phenomena for their operation; Col 22 lines 11-18 The performance metric analysis 630 may collect and analyze metrics related to the performance of prior tasks. For example, the speed, accuracy, and/or cost of a particular quantum algorithm may be determined and stored by the performance metric analysis 630. Aggregate performance metrics may be maintained for quantum computing instance types, quantum algorithms, families of algorithms, initial configurations for algorithms, and so on), that aggregates real-time queue status from a plurality of quantum vendors (the quantum computing resources 160 may be hosted in a data center 199 that is not necessarily part of the provider network 190, Col 10 28-30; Examiner notes: communications between external and internal resources), wherein the telemetry plane identifies quantum jobs for all of the quantum processing units (A client 140 may submit a task description 145 that includes information describing one or more tasks that the client seeks to have the task management service 600 oversee, Col 19 48-51; Examiner notes: the task managements service is aware of all client quantum tasks via the task description);
determining characteristics for the quantum jobs represented in the queue statuses included in the telemetry plane (Col 19 Lines 48-55 A client 140 may submit a task description 145 that includes information describing one or more tasks that the client seeks to have the task management service 600 oversee. The task description 145 may include the name of a quantum algorithm, a description of a problem domain associated with quantum computing, a quantum algorithm itself, high-level programming code, an execution plan, an indication of desired results, and/or other suitable elements), wherein the characteristics include runtime characteristics predicted by an estimator engine (The quantum algorithm testing 1140 may collect, generate, and/or analyze metrics related to the performance, accuracy, and/or cost of the different resources or different algorithms, Col 32 6-9) and quantum vendor-specific quantum processing unit characteristics including qubit error rates, gate speeds, and entanglement limits (Different instance types may have different quantum computing characteristics such as the number or configuration of quantum bits, Col 4 33-35; A quantum computing resource may be selected based (at least in part) on characteristics of the quantum algorithm, e.g., to select a quantum resource that is particularly suited for a specific problem domain such as molecular modeling, traffic optimization, and so on. A quantum computing resource may be selected based (at least in part) on optimization of speed, accuracy, and/or cost, Col 5 35-42; Examiner notes: some quantum resources are better suited for some clients over others via metrics of accuracy, speed, and qubit count);
evaluating user intents associated with the quantum jobs, the characteristics, and vendor criteria to generate placement recommendations for each of the quantum jobs (Col 8 Line 58- Col 9 Line 10 the elastic quantum computing service 100 may include a component for quantum instance recommendation 102. The quantum instance recommendation 102 may recommend the quantum computing instance 161, quantum algorithm 165, and/or initial configuration 166 to the client. For example, based on a particular problem domain or workload that the client specifies, the quantum instance recommendation 102 may recommend a particular instance type of the quantum instance 161 that is designed for the problem domain (vendor criteria). The recommendation may seek to optimize a speed, accuracy, or cost of the problem, based (at least in part) on input concerning the client's goals (user intents). The instance type may be associated with a quantum computing resource that has particular quantum computing characteristics, such as a particular number of qubits. The recommended instance may be pre-loaded with a suitable quantum algorithm 165 and/or an initial configuration 166 (e.g., initial values for the qubits) for that algorithm, or the algorithm may be recommended separately; Based on the results and/or metrics, the environment may recommend one or more particular resource types for a particular quantum algorithm or may recommend a variant of the quantum algorithm, Col 32 9-12);
assigning the quantum jobs to the quantum processing units associated with the placement recommendations (Col 9 lines 13-16 the recommended instance may be provisioned and attached to a classical instance specified by the client based on user input accepting the recommendation; Col 4 line 64-67 the task management service may use both quantum computing resources and classical computing resources to perform a particular task, e.g., where the different types of resources perform different portions of the task); and
orchestrating the execution of the quantum jobs at the assigned quantum processing units, wherein the quantum jobs are executed at the assigned quantum processing units (Col 4 lines 48-51 The task management service may receive task descriptions from clients and select and orchestrate computing resources, including quantum computing resources, to perform the tasks on behalf of the clients).
While Richardson teaches quantum resource recommendation and allocation for quantum tasks, it does not explicitly teach consolidating a queue status of each job queue associated with quantum processing units in a telemetry plane; and wherein each of the job queues includes quantum jobs for corresponding quantum processing units;
However, Im teaches consolidating a queue status of each job queue associated with processing units ([0053] The work queue monitor 140 may periodically monitor the status of each of the work queues 220, 320, and 420 of the job processors 200, 300, and 400. The status information of the work queue may include, for example, the number of stages that are stored in the work queue, stage starting time, time elapsed for executing the stage, and the overall or average time elapsed for executing all stages stored in the work queue. The work queue monitor 140 may provide the status information of the work queues 220, 320 and 420 to the work scheduler 130. Accordingly, the work scheduler 130 may refer to the status information when allocating the stages to the job processors 200, 300, and 400); and wherein each of the job queues includes quantum jobs for corresponding quantum processing units (Fig 2 First Work Queue 220, Second Work Queue 320, Third Work Queue 420; [0044] The respective first, second, and third work queues 220, 320, and 420, store information of the stages that are to be processed in the corresponding first, second, and third cores 210, 310, and 410).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined Im’s work queue monitoring system with the quantum resource allocation system of Richardson resulting in the combined system having work queues for processors to store incoming tasks with the capability to monitor and collect information about the tasks in the queues. A person of ordinary skill in the art would have been motivated to make this combination to provide Richardson’s system with the advantage of staging incoming workloads for analysis to improve resource allocation (see Im [0006] To improve performance of software that includes a large amount of data to be processed, the software may be executed using multiple cores in a parallel manner. In this example, the task to be processed is divided into a plurality of jobs (or stages). The jobs include data and each job is allocated to a specified core for data processing. A static scheduling method may be used to process the plurality of jobs. In the static scheduling method the task to be processed is divided into a number of data segments (jobs) equivalent to the number of cores and jobs are allocated to the cores based on the result of the divided data; [0007] The above described methods use individual work queues for the respective cores, and in each of the methods, the entire data is divided into several segments (jobs) and each segment is allocated to the work queue of a specified core at the beginning of a data process).
Richardson and Im do not explicitly teach wherein the assigning the quantum jobs includes moving a first quantum job in the telemetry plane and associated with a first job queue of a first quantum processing unit of a first quantum vendor of the plurality of quantum vendors to a second job queue of a second quantum processing unit of a second quantum vendor of the plurality of quantum vendors.
However, Ravi teaches wherein the assigning the quantum jobs includes moving a first quantum job in the telemetry plane and associated with a first job queue of a first quantum processing unit of a first quantum vendor of the plurality of quantum vendors to a second job queue of a second quantum processing unit of a second quantum vendor of the plurality of quantum vendors ([0075] jobs 122 may be dynamically moved between different job queues 120 or assigned to different quantum computing devices 132 (e.g., based on availability/expected queuing times, satisfying specific metrics such as maximizing overall machine utilization/effective quantum volume/fidelity, and so forth). In the case a job 122 is dynamically reassigned to a different queue 120 or device 132, its compiled quantum circuit may be re-optimized for the new target device 132. This can again be performed with in-queue optimizations; [0023] Example quantum computing devices include “IBM Q” devices from International Business Machines (IBM), “Bristlecone” quantum computing device from Google, “Tangle Lake” quantum computing device from Intel, and “2000Q” from D-Wave).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined Ravi’s quantum job transferring method with the system of Richardson and Im to transfer/reassign quantum jobs to better quantum resources’ queues. A person of ordinary skill in the art would have been motivated to make this combination to provide Richardson and Im’s system with the advantage of reassigning tasks to better resources for improved system performance in multi-processor systems (see Ravi [0075]).
While Richardson discloses analyzing quantum resource test metrics and providing quantum resource recommendations according to said metrics (Col 32), Richardson, Im, Ravi do not explicitly teach evaluating user intents associated with the quantum jobs, the characteristics, and vendor criteria in combination with a scoring mechanism, to generate placement recommendations for each of the quantum jobs.
However, Lee teaches evaluating user intents associated with the quantum jobs, the characteristics, and vendor criteria in combination with a scoring mechanism, to generate placement recommendations for each of the quantum jobs ([0050] Specifically, the hardware selection module 117 may assign a matching score to each of the plurality of quantum computing hardware according to similarity with resource-related information. In an exemplary embodiment, the hardware selection module 117 may assign a matching score to each of the plurality of quantum computing hardware according to the degree of similarity with the specification of each of the plurality of quantum computing hardware for each item of resource-related information such as the number of qubits, the qubit lifetime, the gate fidelity, the gate operation time, and the qubit connectivity).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined Lee’s quantum resource scoring system with the existing system. A person of ordinary skill in the art would have been motivated to make this combination to provide the resulting system with the advantage of selecting the best suited quantum resource for the task at hand (see Lee [0011] selecting, as quantum computing hardware which is to execute the corresponding standard quantum computing code, the quantum computing hardware with a highest total matching score for each resource item among the plurality of quantum computing hardware.).
Regarding claim 2, Richardson, Im, and Ravi teach the method of claim 1.
Im further teaches receiving the queue status from each of the job queues ([0053] The work queue monitor 140 may provide the status information of the work queues 220, 320 and 420 to the work scheduler 130).
Regarding claim 3, Richardson, Im, and Ravi teach the method of claim 1.
Richardson further teaches wherein the user intents include one or more of execution deadline, accuracy, confidence, budget, and QPU type (Col 9 lines 1-3 The recommendation may seek to optimize a speed, accuracy, or cost of the problem, based (at least in part) on input concerning the client's goals; Col 35 lines 39-41 user-specified goals related to resource usage, budgetary constraints, runtime, performance, and so on).
Regarding claim 4, Richardson, Im, and Ravi teach the method of claim 1.
Richardson further teaches wherein the vendor criteria include one or more of circuit requirements, real-time quantum processing unit telemetry, and runtime characteristics (Col 8 Line 63- Col 9 line 9 based on a particular problem domain or workload that the client specifies, the quantum instance recommendation 102 may recommend a particular instance type of the quantum instance 161 that is designed for the problem domain. The recommendation may seek to optimize a speed, accuracy, or cost of the problem, based (at least in part) on input concerning the client's goals. The instance type may be associated with a quantum computing resource that has particular quantum computing characteristics, such as a particular number of qubits. The recommended instance may be pre-loaded with a suitable quantum algorithm 165 and/or an initial configuration 166 (e.g., initial values for the qubits) for that algorithm; Examiner notes: the recommender weighs the provider resource’s hardware configuration type, # qubits, runtime speed, cost, accuracy to make a recommendation of quantum resource).
Regarding claim 7, Richardson, Im, and Ravi teach the method of claim 1. Richardson teaches quantum jobs (Col 25 lines 54-60 a task description that includes information describing one or more tasks that the client seeks to have the task management service oversee. The task description may include the name of a quantum algorithm, a description of a problem domain associated with quantum computing, a quantum algorithm itself) and quantum processing units (Col 6 lines 38-40 Quantum computing resources 160 may include computing devices that rely on quantum-mechanical phenomena for their operation).
Im further teaches splitting at least one of the quantum jobs into multiple circuits and assigning the multiple circuits to one or more of the quantum processing units ([0040] the device processors 200, 300, and 400 may perform pipeline processing of an application in parallel and in time-sliced fashion by dividing the application into two or more stages and executing the application by stages. In this example, the first through third device processors 200, 300, and 400 execute stages allocated to them and under the control of the host processor 100).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined Im’s parallel processing of stages of an application with the quantum resource allocation system of Richardson resulting in quantum resources able to process quantum tasks in parallel. A person of ordinary skill in the art would have been motivated to make this combination to provide Richardson’s system with the advantage of accelerating task processing through the use of parallel processing (see Im [0006] To improve performance of software that includes a large amount of data to be processed, the software may be executed using multiple cores in a parallel manner).
Regarding claim 8, Richardson, Im, and Ravi teach the method of claim 7.
Richardson further teaches aggregating results generated from executing the multiple circuits (Col 16 lines 14-17 The library 400 may forward the results 117 directly to the application 115 or may aggregate the results of multiple runs before forwarding the aggregated results to the application).
Regarding claim 9, Richardson, Im, and Ravi teach the method of claim 1.
Richardson further teaches optimizing the placement recommendations and reassigning the quantum jobs to the quantum processing units (Col 28 lines 49-56 the optimization may use metrics from prior runs to improve the speed, cost, and/or accuracy of the program code, e.g., by structuring quantum gates in a different way. In one embodiment, the optimization may be performed relative to a particular type of quantum computing resource, e.g., a quantum computer with a particular number of quantum bits or a particular hardware configuration; Examiner notes: changing the underlying quantum resources is reassigning the tasks to a different set of quantum processing units).
Regarding claim 10, Richardson, Im, and Ravi teach the method of claim 1.
Richardson further teaches wherein the quantum processing units include one or more of on-premise quantum processing units (Col 12 lines 2-4 the quantum computing instance 161 may be respectively provisioned within enterprises having their own internal networks), edge based quantum processing units (Col 11 lines 54-58 The classical computing instances may be located in different data centers or different physical environments than the quantum computing resources 160, and a network 150 may connect the classical instances to the quantum instances), and cloud based quantum processing units (Col 18 lines 49-52 the quantum computing resources 160 may be hosted “in the cloud,” e.g., in a cloud-based provider network 190 whose resources are remotely accessible to clients), wherein types include physical quantum processing units (Col 19 lines 17-19 one quantum computing instance may correspond to one physical quantum computing device in the quantum resources 160) and virtual quantum processing units (Col 19 lines 30-32 a quantum computing instance may represent a virtual instance that is implemented on top of the underlying physical quantum resources 160).
Regarding claim 11, it is the non-transitory storage medium of claim 1. Therefore, it is rejected for the same reasons as claim 1.
Richardson further teaches a non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations (Col 40 lines 42-45 A non-transitory computer-readable storage medium may also include any volatile or non-volatile media such as RAM (e.g. SDRAM, DDR SDRAM, RDRAM, SRAM, etc.), ROM, etc.,).
Regarding claims 12-14 and 17-20, they are the non-transitory storage media of claims 2-4 and 7-10. Therefore, they are rejected for the same reasons as claims 2-4 and 7-10.
Claims 6 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Richardson et al. US 11270220 B1 in view of Im et al. US 20110161965 A1 in view of Ravi et al. US 20240378085 A1 in view of Lee et al. US 20210357278 A1 in further view of Griffin et al. US 20220383171 A1 (“Griffin”).
Griffin is cited in a previous office action.
Regarding claim 6, Richardson, Im, and Peacock teach the method of claim 1.
Richardson teaches assigning the quantum jobs based on entanglement characteristics of the quantum processing units that match the quantum jobs (Col 8 line 65 – Col 9 line 9 the quantum instance recommendation 102 may recommend a particular instance type of the quantum instance 161 that is designed for the problem domain. The recommendation may seek to optimize a speed, accuracy, or cost of the problem, based (at least in part) on input concerning the client's goals. The instance type may be associated with a quantum computing resource that has particular quantum computing characteristics, such as a particular number of qubits. The recommended instance may be pre-loaded with a suitable quantum algorithm 165 and/or an initial configuration 166 (e.g., initial values for the qubits) for that algorithm; Col 9 lines 13-16 the recommended instance may be provisioned and attached to a classical instance specified by the client based on user input accepting the recommendation).
Richardson, Im, Ravi, and Lee do not explicitly teach assigning the quantum jobs based on entanglement characteristics of the quantum processing units that match the quantum jobs.
However, Griffin teaches assigning the quantum jobs based on entanglement characteristics of the quantum processing units that match the quantum jobs ([0018] If the one or more qubits are available, the QBS allocates the one or more qubits for the executing quantum service (e.g., by invoking functionality of the qubit registry, as a non-limiting example). However, if the QBS determines that the one or more qubits are unavailable for allocation, the QBS places the executing quantum service into a sleep state (e.g., by invoking functionality of a quantum service manager, as a non-limiting example); Claim 21: determining the count of the number of available qubits that are currently available for allocation comprises, for at least one qubit of the total number of qubits, evaluating an entanglement indicator for the at least one qubit, wherein the entanglement indicator is indicative of whether the at least one qubit is in an entangled state).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to have combined Griffin’s checking of qubit availability based on an entanglement indicator with the quantum resource allocation system of Richardson, Im, Ravi, and Lee. A person of ordinary skill in the art would have been motivated to make this combination to provide Richardson system with the advantage of being able to better monitor the qubit availability of quantum resources so that the optimal resource allocations for tasks requiring non entangled qubits are able to be allocated quicker (see Griffin [0015] Because qubits generally require very specific environmental conditions for operation, the number of qubits available to quantum services that are executing on a given quantum computing device may be limited. Moreover, if qubits are assigned specific properties during execution (e.g., placed in a state of entanglement or superposition, as non-limiting examples) by a first quantum service, an attempt by a second quantum service to allocate or access those qubits may result in a disruption of those specific properties, which may cause operations of the first quantum service to fail. Thus, an ability to efficiently manage runtime qubit allocation for executing quantum services will be desirable).
Regarding claim 16, it is the non-transitory storage medium of claim 6. Therefore, it is rejected for the same reasons as claim 6.
Conclusion
Any inquiry concerning this communication or earlier communications from the
examiner should be directed to HARRISON LI whose telephone number is (703) 756-1469. The
examiner can normally be reached Monday-Friday 9:00am-5:30pm ET.
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, Aimee Li can be reached on (571) 272-4169. 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.
/H.L./
Examiner, Art Unit 2195
/Aimee Li/Supervisory Patent Examiner, Art Unit 2195