Patent Application 17992684 - LEARNING MODEL TO OPTIMIZE AUTOMATIC WORKLOAD - Rejection

Patent Application 17992684 - LEARNING MODEL TO OPTIMIZE AUTOMATIC WORKLOAD

Title: LEARNING MODEL TO OPTIMIZE AUTOMATIC WORKLOAD MIGRATION

Application Information

• Invention Title: LEARNING MODEL TO OPTIMIZE AUTOMATIC WORKLOAD MIGRATION
• Application Number: 17992684
• Submission Date: 2025-05-14T00:00:00.000Z
• Effective Filing Date: 2022-11-22T00:00:00.000Z
• Filing Date: 2022-11-22T00:00:00.000Z
• National Class: 718
• National Sub-Class: 108000
• Examiner Employee Number: 86765
• Art Unit: 2194
• Tech Center: 2100

Rejection Summary

• 102 Rejections: 1
• 103 Rejections: 0

Cited Patents

The following patents were cited in the rejection:

• US 0153896🔗

Office Action Text

    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
The instant application having Application No. 17/992,684 filed on 11/22/2022 is presented for examination.
Examiner Notes
Examiner cites particular columns and line numbers in the references as applied to the claims below 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.

Drawings
The applicant’s drawings submitted are acceptable for examination purposes.

Authorization for Internet Communications
The examiner encourages Applicant to submit an authorization to communicate with the examiner via the Internet by making the following statement (from MPEP 502.03):
“Recognizing that Internet communications are not secure, I hereby authorize the USPTO to communicate with the undersigned and practitioners in accordance with 37 CFR 1.33 and 37 CFR 1.34 concerning any subject matter of this application by video conferencing, instant messaging, or electronic mail. I understand that a copy of these communications will be made of record in the application file.”

Please note that the above statement can only be submitted via Central Fax, Regular postal mail, or EFS Web.

Claim Rejections - 35 USC § 102
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)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.

Claims 1-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Kumar (US 2020/0153896).

As per claim 1, Kumar discloses a method, comprising:
monitoring, via a communication network, by a first computing system comprising a processor, a first metric corresponding to a second computing system of a set of computing systems that are available to provide computing services via the communication network (Fig. 3, 310);
analyzing, by the first computing system, under control of a first trained learning model, the first metric with respect to a first metric migration criterion, the analyzing resulting in an analyzed first metric (Fig. 3, 320);
generating, by the first computing system, under control of the first trained learning model, first migration data representative of a first migration determination to migrate a first workload from the second computing system to a third computing system of the set of computing systems based on the analyzed first metric indicating that the first metric satisfied the first metric migration criterion (Fig. 3, 350); and
responsive to the first migration determination, causing, by the first computing system, migration of the first workload from the second computing system to the third computing system (Fig. 3, 360).

As per claim 2, Kumar further discloses further comprising:
monitoring, via the communication network, a second metric corresponding to the third computing system of the set of computing systems, wherein the first metric migration criterion is evaluated based at least in part on the second metric (Paragraph 53 “In step 360, intelligent optimizer 110 may select the execution plan that has the lowest processing cost (e.g., processor utilization demand). The selected execution plan may be later used by server cluster 108 to process the workload.”).

As per claim 3, Kumar further discloses wherein training of the first trained learning model comprises:
generating migration recommendation data representative of migration recommendations of whether to migrate one or more training computing workloads;
responsive to the migration recommendations, receiving migration selection data representative of respective migration selections corresponding to the migration recommendations; and
updating an initial learning model based on the respective migration selections and migration recommendations to which the respective migration selections correspond, the updating resulting in the first trained learning model (Paragraph 350 “In step 350, intelligent optimizer 110 may determine costs of processing the workload by server cluster 108 under one or more alternative execution plans. Because the current resource availability usually cannot be changed, intelligent optimizer 110 may provide the alternative execution plans to determine whether the cost of processing the workload can be lowered.).

As per claim 4, Kumar further discloses wherein at least one of the migration selections comprises an instruction to migrate at least one of the one or more training computing workloads, wherein the training of the first trained learning model further comprises:
analyzing a first migration selection of the migration selections with respect to a first migration recommendation of the migration recommendations according to a migration recommendation acceptance function, the analyzing of the first migration selection resulting in a first determined migration recommendation acceptance;
analyzing a second migration selection of the migration selections with respect to a second migration recommendation of the migration recommendations according to the migration recommendation acceptance function, the analyzing of the second migration selection resulting in a second determined migration recommendation acceptance; and
updating the first metric migration criterion based on the first determined migration recommendation acceptance or the second determined migration recommendation acceptance (Paragraph 350 “In step 350, intelligent optimizer 110 may determine costs of processing the workload by server cluster 108 under one or more alternative execution plans. Because the current resource availability usually cannot be changed, intelligent optimizer 110 may provide the alternative execution plans to determine whether the cost of processing the workload can be lowered.).

As per claim 5, Kumar further discloses wherein at least one of the first determined migration recommendation acceptance or the second determined migration recommendation acceptance is manually configured (Paragraph 32 “Workload management engine 106 may also include one or more I/O devices 220 that may comprise one or more interfaces for receiving signals or input from devices and providing signals or output to one or more devices that allow data to be received and/or transmitted by workload management engine 106. For example, workload management engine 106 may include interface components, which may provide interfaces to one or more input devices, such as one or more keyboards, mouse devices, and the like, that enable workload management engine 106 to receive input from a user or administrator (not shown).”).

As per claim 6, Kumar further discloses wherein at least one of the migration selections comprises an instruction for at least one workload, of the one or more training computing workloads, to remain on a computing system via which the at least one workload is currently executing, wherein the training of the first trained learning model further comprises:
analyzing a first migration selection of the migration selections according to a migration recommendation acceptance function with respect to a first migration recommendation of the migration recommendations corresponding to the first migration selection, the analyzing of the first migration selection resulting in a first determined migration recommendation acceptance;
analyzing a second migration selection of the migration selections according to a migration recommendation acceptance function with respect to a second migration recommendation of the migration recommendations corresponding to the second migration selection, the analyzing of the second migration selection resulting in a second determined migration recommendation acceptance; and
updating the first metric migration criterion based on at least one of the first determined migration recommendation acceptance or the second determined migration recommendation acceptance (Fig. 3, step 350).

As per claim 7, Kumar further discloses wherein the initial learning model is configured with factors of a static migration rules engine (Paragraph 50 “In step 330, intelligent optimizer 110 may use the machine learning algorithms to construct a cost model for predicting the time and resources consumed by a workload. Intelligent optimizer 110 may construct the cost model based at least on the hardware capacities and correlations of utilizations determined in step 320. The machine learning algorithms may be any algorithm known in the art, such as the k-nearest neighbor algorithm, the linear regression algorithm, the decision table/tree algorithm, the artificial neural network (ANN), the support vector machine algorithm, time series method, etc. In exemplary embodiments, steps 310-330 may be routinely repeated so that intelligent optimizer 110 may continuously learn the system behavior and refine the cost model.).

As per claim 8, Kumar further discloses wherein the respective migration selections corresponding to the migration recommendations are manually input to the initial learning model or to the first trained learning model (Paragraph 32 “Workload management engine 106 may also include one or more I/O devices 220 that may comprise one or more interfaces for receiving signals or input from devices and providing signals or output to one or more devices that allow data to be received and/or transmitted by workload management engine 106. For example, workload management engine 106 may include interface components, which may provide interfaces to one or more input devices, such as one or more keyboards, mouse devices, and the like, that enable workload management engine 106 to receive input from a user or administrator (not shown).”).

As per claim 9, Kumar further discloses wherein the first workload, when executing via the second computing system, comprises instructions to access data from a data store corresponding to a fourth computing system communicatively coupled with the communication network, and wherein the first migration data representative of the first migration determination comprises instructions for the first workload to access the data from the data store after the first workload has been migrated to the third computing system (Fig. 3, steps 350 and 360).

As per claim 10, Kumar further discloses wherein the first trained learning model corresponds to a first type of learning model for an instance of the first workload being a first type of workload, and wherein the first trained learning model corresponds to a second type of learning model for the instance of the first workload being a second type of workload (Paragraph 8 “automated server workload management using machine learning. In particular, the disclosed systems and methods predict a distributed computing system's efficiency of processing a workload by using machine learning algorithms to study historical behavior of the system.).

As per claim 11, Kumar further discloses further comprising:
monitoring, via the communication network, by the first computing system, a second metric corresponding to a fourth computing system of the set of computing systems;
analyzing, under control of a second trained learning model executing via the first computing system, the second metric with respect to a second metric migration criterion, the analyzing of the second metric resulting in an analyzed second metric;
generating, under control of the second trained learning model, second migration data representative of a second migration determination to migrate a second workload executing via the fourth computing system to a fifth computing system of the set of computing systems based on the analyzed second metric satisfying the second metric migration criterion; and
responsive to the second migration determination, causing, by the first computing system, migration of the second workload from the fifth computing system to the sixth computing system,
wherein the first trained learning model and the second trained learning model are different (Fig. 3, step 350).

As per claim 12, it is a system claim having similar limitations as cited in claim 1 and is thus rejected under the same rationale.

As per claim 13, it is a system claim having similar limitations as cited in claim 9 and is thus rejected under the same rationale.

As per claim 14, it is a system claim having similar limitations as cited in claim 2 and is thus rejected under the same rationale.

As per claim 15, Kumar further discloses wherein the first computing system and the fourth computing system are different computing systems (Paragraph 42).

As per claim 16, it is a system claim having similar limitations as cited in claim 4 and is thus rejected under the same rationale.

As per claim 17, it is a medium claim having similar limitations as cited in claim 1 and is thus rejected under the same rationale.

As per claim 18, it is a medium claim having similar limitations as cited in claims 3 and 4 and is thus rejected under the same rationale.

As per claim 19, Kumar further discloses wherein the first workload metric or the second workload metric corresponds to at least one of: a disruption tolerance, a first cost of the first portion of the first workload executing on the second computing system (Paragraph 29 “The processing cost may be the time and resources consumed by the workload.), a second cost of the first portion of the first workload executing on the third computing system, a third cost of the first portion of the first workload per unit of time on premises, a percentage of time at peak, a virtual disk storage cost associated with the second computing system, a virtual disk storage cost associated with the third computing system, a fourth cost of a first virtual machine associated with the second computing system at a first point in time, a fifth cost of a second virtual machine associated with the third computing system at a second point in time, a sixth cost of a third virtual machine per storage unit of a virtual storage associated with the second computing system, a seventh cost of a fourth virtual machine per storage unit of a virtual storage associated with the third computing system, an eighth cost of a fifth virtual machine per storage unit of a virtual storage associated with the second computing system at a third point in time, a ninth cost of a sixth virtual machine per storage unit of a virtual storage associated with the third computing system at a fourth point in time, a tenth cost of bandwidth to migrate the first portion of the first workload from the second computing system to the third computing system, an eleventh cost to migrate the first portion of the first workload from the second computing system to the third computing system, a first workload mode of steady, or a second workload mode of variable.

As per claim 20, Kumar further discloses wherein the first portion of the first workload, when executing on the second computing system, comprises first instructions to access data from a storage communicatively coupled with the communication network, and wherein the migration determination comprises second instructions for the first portion of the first workload to access the data from the storage communicatively coupled with the communication network after the first portion of the first workload has been migrated to the third computing system (Fig. 3, 360).

Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TIMOTHY A MUDRICK whose telephone number is (571)270-3374. The examiner can normally be reached 9am-5pm Central Time.
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, Kevin Young can be reached at (571) 270-3180. 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.
/TIMOTHY A MUDRICK/Primary Examiner, Art Unit 2194                                                                                                                                                                                                        5/10/2025