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 .
DETAILED ACTION
Claims 1-16 and 21-24 are pending. Claims 17-20 are canceled and claims 21-24 are newly added by Applicant.
Examiner Notes
Examiner cites particular paragraphs and/or columns and lines in the references as applied to Applicant’s claims for the convenience of the Applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the Applicant fully consider the references in entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner. The prompt development of a clear issue requires that the replies of the Applicant meet the objections to and rejections of the claims. Applicant should also specifically point out the support for any amendments made to the disclosure. See MPEP § 2163.06.
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 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.
Request for Continued Examination
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 01/12/2026 has been entered.
Claim Objections
As per claim 22, replace “an average amount of response to process workloads” and “an average amount of work done to process a workload by said executor” with “the average amount of work to execute a workload of said task by said executor”. More specifically, [0059] of the instant specification discloses that the average amount of response to process a workload is equivalent to the average amount of work done which is already being claimed in claim 1 from which claim 22 depends. Therefore, the aforementioned language of claim 22 would be duplicative of that recited in claim 1. 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.
Claims 1, 7-8, and 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino et al. (US 2012/0110585) (hereinafter Cosentino) in view of Eastep et al. (US 2016/0188380) (hereinafter Eastep).
As per claim 1, Cosentino primarily teaches the invention as claimed including a system comprising:
a memory ([0022]); and
a processor in communication with said memory, said processor being configured to execute operations ([0022]) comprising:
receiving a request to execute a task from a user ([0036] execution agent enforces the execution of each job in response to a corresponding request received from the executor), wherein said task is associated with an executor of a plurality of executors ([0036] handler submits each job of the plan for execution as soon as possible on the corresponding execution server either defined statically or selected at run-time among the available execution servers having the required characteristics) based on an optimal amount of power to be used by said executor to execute said task (abstract execution scheme optimizes the energy consumed by the data-processing system to execute the data-processing jobs and claim 13 select a data-processing unit to execute each data-processing job to optimize the energy consumed according to the ambient temperature of the plurality of data-processing units);
executing said task with said executor ([0036] execution agent enforces the execution of each job in response to a corresponding request received from the executor); and
returning a response to said request to said user ([0036] return feedback information indicating the result of the execution which may include indicating whether the job has completed successfully, the actual duration of the job, or the like).
Cosentino does not explicitly teach provisioning an executor with resources to fulfill said request based on an average amount of work to execute a workload of said task by said executor task.
However, Eastep teaches provisioning an executor with resources to fulfill said request based on an average amount of work to execute a workload of said task by said executor task ([0060]-[0061] progress meters of nodes may report average work completed and/or average work to be completed across cores of the nodes and based on that information the allocation of resources applied to the nodes may be appropriately modified).
Eastep and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep because to it would provide a way to reduce total waiting times for task execution by reallocating computational resources at work on a task. Waiting times at a core level may be reduced overall by speeding up slower cores while slowing down faster cores to allow threads and/or cores to arrive at a global synchronization barrier in relatively less mean time. Core (and/or processor) frequency may be varied over a range by altering the amount of power that may be fed to the core (and/or processor). In a situation where power resources may be limited, faster thread processing times may be obtained by shifting power away from cores that are faster than the average of the cores employed, and toward cores that are slower than the average of the cores employed. In some circumstances, it may be advantageous to redirect power away from cores that are slower than average to other cores that are even slower. Progress meters provide data that may be used to regularly adjust power to cores, thereby relatively reducing waiting times at synchronization points. Power shifting may also reduce power consumed in the course of processing a given job. Advantageously, the overall speed with which a parallel processing job is to be completed may be relatively increased. In addition, total amount of power necessary to complete the job may be relatively reduced. Thus, reducing the relative power consumed by the cores may result in less heat generated at the cores, which may allow relatively less intensive use of air conditioning systems in facilities to provide additional further power savings.
As per claim 7, Cosentino further teaches wherein the response comprises one or more of a calculation, a search result, a computation, an answer to a question, and a confirmation of a success ([0036] return feedback information indicating the result of the execution which may include indicating whether the job has completed successfully, the actual duration of the job, or the like).
As per claim 8, it has similar limitations as claim 1 and is therefore rejected using the same rationale.
As per claim 14, it has similar limitations as claim 7 and is therefore rejected using the same rationale.
As per claim 15, Eastep teaches wherein said executor comprises a plurality of nodes ([0013]-[0014] cluster/group of nodes).
As per claim 16, it has similar limitations as claim 1 and is therefore rejected using the same rationale.
Claims 2 and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Thomas (US 10,616,078) (as previously cited).
As per claim 2, Consentino in view of Eastep do not explicitly teach obtaining usage data associated with a plurality of executors, wherein said usage data is associated with a usage pattern.
However, Thomas teaches obtaining usage data associated with a plurality of executors, wherein said usage data is associated with a usage pattern (fig. 2, blocks 210-250 monitor compute resources usage, store usage data in a data store, analyze stored usage data to determine a usage pattern of the resources).
Thomas and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Consentino in view of Eastep in view of Thomas because it would provide for a profiling service that can use pre-determined baseline usage patterns for a type of resource. For example, the baseline usage pattern for a type of resource can be supplied through a database having fields populated by an administrator of the distributed environment. The pre-determined baseline usage patterns for a type of resource can be based on historical usage patterns or desired usage patterns. Using pre-determined baseline usage patterns for a type of resource may reduce the computing overhead associated with frequently calculating running baseline usage patterns for a type of resource.
As per claim 9, it has similar limitations as claim 2 and is therefore rejected using the same rationale.
Claims 3 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Thomas in view of Desikachari et al. (US 2015/0310139) (Desikachari as previously cited).
As per claim 3, Cosentino in view of Eastep in view of Thomas do not explicitly teach wherein said executor is associated with a resource selection model, and wherein said resource selection model is based on said usage pattern.
However, Desikachari teaches wherein said executor is associated with a resource selection model, and wherein said resource selection model is based on said usage pattern (fig. 4 and [0022]-[0023] generate a model for a resource based on a selected set of usage patterns).
Desikachari and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Thomas in view of Desikachari because it would provide for a way of intelligently determining approximate resource consumptions by various transactions comprising an application and scaling infrastructure components accordingly based on maximum expected load which results in a more optimal allocation of resources for implementing the application. This may be done with respect to adding resources to facilitate application growth, and may be similarly employed to remove extra or unnecessary resources to decrease application capacity if desired.
As per claim 10, it has similar limitations as claim 3 and is therefore rejected using the same rationale.
Claims 4 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Thomas in view of Cronin (US 2023/0211691) (as previously cited).
As per claim 4, Cosentino in view of Eastep in view of Thomas do not explicitly teach wherein a lifecycle of said executor is based on said usage data.
However, Cronin teaches wherein a lifecycle of said executor is based on said usage data ([0166] build different ML models based on system usage data to extend the lifecycle).
Cronin and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Thomas in view of Cronin because it would provide a way for pre-allocating slots in hosts for a particular user account what may not be deemed necessary to ensure host availability for that user account and, reserving the host slots would otherwise waste resources as no other user accounts could use the slots. Alternatively, rather than not creating the allocation upon determining that the fleet of hosts is likely to be widely available for executing virtual machines, a capacity management system may create the allocation but set the start time later rather than earlier to reduce the amount of time that the slots are unavailable for use by other users. Upon expiration of the end time, the allocation is considered invalid and the allocated slots in the specified hosts can be used to launch virtual machines for any user accounts.
As per claim 11, it has similar limitations as claim 4 and is therefore rejected using the same rationale.
Claims 5 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep, and further in view of Ramaswamy et al. (US 2023/0081488) (hereinafter Ramaswamy as previously cited).
As per claim 5, Cosentino in view of Eastep do not explicitly teach wherein a type of said executor type is one of a serverless type, a microservice type, and a mixed service type.
However, Ramaswamy teaches wherein a type of said executor type is one of a serverless type, a microservice type, and a mixed service type ([0034] a backend can be a mix of Microservices and Serverless functions).
Ramaswamy and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Ramaswamy because it would provide for a canvas system which can be optimized for scale and development speed. The canvas system can use different kinds of databases that are the best fitting for the specified use cases and use an amalgamation of all of these to get the best performance at scale while still being cost-effective.
As per claim 12, it has similar limitations as claim 5 and is therefore rejected using the same rationale.
Claims 6 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Gupta et al. (US 11,500,663) (hereinafter Gupta as previously cited).
As per claim 6, Cosentino in view of Eastep do not explicitly teach converting said executor from a first executor type to a second executor type.
However, Gupta teaches converting said executor from a first executor type to a second executor type (col. 7, ll. 17-32 reconfiguring servers to execute different types of virtual machines).
Gupta and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Gupta because it would provide a way for pre-allocating slots in hosts for a particular user account what may not be deemed necessary to ensure host availability for that user account and, reserving the host slots would otherwise waste resources as no other user accounts could use the slots. Alternatively, rather than not creating the allocation upon determining that the fleet of hosts is likely to be widely available for executing virtual machines, a capacity management system may create the allocation but set the start time later rather than earlier to reduce the amount of time that the slots are unavailable for use by other users. Upon expiration of the end time, the allocation is considered invalid and the allocated slots in the specified hosts can be used to launch virtual machines for any user accounts.
As per claim 13, it has similar limitations as claim 6 and is therefore rejected using the same rationale.
Claim 21 is rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Blanding et al. (US 9,515,905) (hereinafter Blanding) in view of Karri et al. (US 2023/0071278) (hereinafter Karri).
As per claim 21, Cosentino in view of Eastep do not explicitly teach wherein said optimal amount of power is based on a user priority, wherein said user priority is selected from the group consisting of minimizing resource consumption, minimizing an execution cost, and using minimal resources to complete said task within a defined period of time.
However, Blanding teaches wherein said optimal amount of power is based on a user priority (col. 4, ll. 66 to col. 5, ll. 7 specify relative priorities of workloads and execution environments), wherein said user priority is selected from the group consisting of minimizing resource consumption, minimizing an execution cost (col. 3, ll. 23-28 minimize resource cost and energy consumption).
Blanding and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Blanding because it would provide for automatic management of multiple scale out workloads so as to minimize the computing resources and hence the related energy consumption required to achieve the performance targets established for the workloads. Automatic management provides for reduced cost and better matching of resources to actual demand as compared to manual management. Automatic management of heterogeneous resources (e.g., use of both virtual and physical execution environment) provides more flexibility than managing only homogeneous resources (e.g., only virtual machines) as well as potential for a better match between required and available resources. Automatic management of heterogeneous workloads from a single resource pool provides more efficiency in use of computing resources as opposed to using separate pools for each workload, and thus reduces energy consumption by the computer system.
Cosentino in view of Eastep in view of Blanding do not explicitly teach using minimal resources to complete said task within a defined period of time.
Howerver, Karri teaches using minimal resources to complete said task within a defined period of time ([0013] optimization criteria seeks to minimize the amount of computational resources to allocate to process the execution paths within required available times to execute the activity steps).
Karri and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Blanding in view of Karri because it would provide for an optimization criteria to minimize execution time by selecting a group of execution paths that completes execution in a minimum time. If the optimization criteria is to minimize processor/memory resource allocation to conserve computational resources, then the resource allocation selects the group of execution paths that minimizes the processor/memory resource allocation and completes execution of the activity steps in the required available times for the activity steps.
Claim 22 is rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Gmach et al. (US 2013/0268940) (hereinafter Gmach) in view of Fellenstein et al. (US 2006/0149714) (hereinafter Fellenstein).
As per claim 22, Eastep teaches wherein said provisioning said executor with resources to fulfill said request is based on a factor selected from the group consisting of an average amount of response to process workloads, an average amount of work done to process a workload by said executor ([0060]-[0061] progress meters of nodes may report average work completed and/or to be completed across cores of the nodes and based on that information the allocation of resources applied to the nodes may be appropriately modified).
Cosentino in view of Eastep do not explicitly teach an average cost for workload services, and a total cost for all available workload services.
However, Gmach teaches an average cost for workload services ([0040] a best average effective cost for each workload using a cloud service to be serviced by virtual machines).
Gmach and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Gmach because it would allow a service user i.e., a customer to decide how best to have workloads hosted by apportioning costs that are least sensitive to workload placement decisions and by providing robust and repeatable cost estimates. The resulting system could compare the costs of hosting a workload in virtualized and non-virtualized environments, separate workloads into categories including those that should be virtualized and those that should not, and determine the amount of physical resources to cost-effectively host a set of workloads.
Cosentino in view of Eastep in view of Gmach do not explicitly teach a total cost for all available workload services.
However, Fellenstein teaches a total cost for all available workload services ([0081] determine the total hardware and software costs for use of the available grid resources for the grid job and to identify the lowest cost resources available).
Fellenstein and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Gmach in view of Fellenstein because it would provide for a grid catalog and storage subsystem of grid services that manages the storage and distribution of software images for efficient resource building facilitates grid services such that a user may specify preferred performance levels for a resource management service to distribute jobs to maintain preferred performance levels within the grid execution environment. A grid dynamic build service can build the resource nodes identified as the most cost effective.
Claim 23 is rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Stefani et al. (US 11,170,309) (hereinafter Stefani).
As per claim 23, Cosentino in view of Eastep do not explicitly teach wherein said provisioning said executor with said resources to fulfill said request is responsive to said executor having inadequate available resources to execute said task.
However, Stefani teaches wherein said provisioning said executor with said resources to fulfill said request is responsive to said executor having inadequate available resources to execute said task (col. 12, ll. 37-48 provision additional computing resources when processing resources available for use drops below a threshold value).
Stefani and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Stefani because it would provide for an efficient machine learning model inference routing system in a machine learning service feedback processing system that collectively allow for adjusting the routing of inferences based on machine learning model accuracy, performance, demand, and/or the like. This could enable the performance of shadow testing, support ensemble machine learning models, and improve existing machine learning models using feedback data.
Claim 24 is rejected under 35 U.S.C. 103 as being unpatentable over Cosentino in view of Eastep in view of Tan et al. (US 2018/0032375) (hereinafter Tan).
As per claim 24, Cosentino in view of Eastep do not explicitly teach wherein said executor is selected to execute said task based on an impact of said executor executing said task.
However, Tan teaches wherein said executor is selected to execute said task based on an impact of said executor executing said task ([0020] when multiple computing frameworks can execute a same sub-task, a target computing framework is selected from the multiple computing frameworks according to operation time and resource consumption, to execute a sub-task, so as to improve the data processing efficiency and working performance of a system).
Tan and Cosentino are both concerned with task execution in computing environments and are therefore combinable/modifiable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cosentino in view of Eastep in view of Tan because when multiple computing frameworks can execute a same task, a target computing framework can be selected from the multiple computing frameworks according to operation time and resource consumption, to execute a sub-task, so as to improve the data processing efficiency and working performance of the system. APIs that are in different computing frameworks and that execute a same type of task are encapsulated using a preset programming language, to generate a unified API, so as to shield differences between programming languages, greatly reduce a programming threshold for a developer, and improve flexibility and adaptability of each computing framework.
Response to Arguments
Applicant’s arguments with respect to the 35 U.S.C. 103 prior art rejections have been considered but are moot because the new ground of rejection necessitated by Applicant’s amendments does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Citation of Relevant Prior Art
The prior art made of record and not relied upon is considered pertinent to Applicant's disclosure:
Ueda (US 2012/0254443) disclose scaling up resources based on an average value of response performance triggering the scale-up.
Jacobson et al. (US 2014/0282605) disclose a data processing engine may be selected upon determining that the data processing engine best satisfies predefined criteria, such as one or more of speed, efficiency, resource consumption, job execution success rate, user-specified execution time constraints, etc.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Adam Lee whose telephone number is (571) 270-3369. The examiner can normally be reached on M-TH 8AM-5PM.
If attempts to reach the above noted Examiner by telephone are unsuccessful, the Examiner’s supervisor, Pierre Vital, can be reached at the following telephone number: (571) 272-4215. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated-interview-request-air-form.
/Adam Lee/Primary Examiner, Art Unit 2198 January 26, 2026