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 .
Status of the Claims
Claims 1, 11, and 21 have been amended. Claims 6-10 and 16-20 have been canceled. Claims 26-35 are new. Claims 1-5, 11-15, and 21-35 are currently pending and have been considered by the Examiner.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-5, 11-15, and 21-35 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the activity logs" in line 14. There is insufficient antecedent basis for this limitation in the claim. Examiner treats this limitation as “activity logs”.
Claims 2-5 and 26-29 are rejected for failing to cure the deficiencies of claim 1.
Claim 11 is rejected because it recites the same indefinite limitation as claim 1. Claims 12-15 and 30-32 are rejected for failing to cure the deficiencies of claim 11.
Claim 21 is rejected because it recites the same indefinite limitation as claim 1. Claims 22-25 and 33-35 are rejected for failing to cure the deficiencies of claim 21.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-5, 11-15, 21-26, 29, 32, and 35 are rejected under 35 U.S.C. 103 as being unpatentable over Zolotow et al. (US 20200019311 A1, cited in PTO-892 issued 06/20/2025) in view of Tan (US 20100169253 A1, cited in PTO-892 issued 06/20/2025) and Nagpal et al. (US 20220156114 A1).
Regarding claim 1, Zolotow teaches: A method of training a machine learning algorithm for identifying an edge computing environment location as a target for workload migration, the method comprising: ([0042]-[0043]. The claim limitation “a machine learning algorithm” comprises machine learning module 118 in combination with rules engine 124. The remaining claim limitations from the preamble are addressed below.)
…
receiving: i) a description of a training workload comprising
receiving ii) an identifier of an edge location for migrating the training workload; ([0038], lines 1-7, [0043], lines 17-21 and Fig. 1 disclose receiving a desired optimal storage environment 126 during training. Each storage 106 may be identified by a number, and they are edge locations because they are at the edge of the network 108 as depicted in Fig. 1.)
training the machine learning algorithm based on the description of a training workload as a training input for the machine learning algorithm and the identifier of the edge location for migrating the training workload as a training output for the machine learning algorithm; and ([0043])
However, Zolotow does not explicitly teach: deploying the machine learning algorithm at the edge computing environment location;
receiving: i) a description of a training workload comprising a set of virtual machines (VMs) and a set of metadata for different applications to migrate;
in response to migrating a workload to the edge computing environment location that is identified based on an output of the trained machine learning algorithm, receiving feedback on performance of the machine learning algorithm by tracking migration metrics in the activity logs at the edge computing environment location.
But Tan teaches: deploying the machine learning algorithm at the edge computing environment location; ([0026], [0034], line 1, and [0035], lines 1-6 disclose an embodiment in which host #1 is a server and host #2 is a personal computer (PC) comprising ANNs, and the PC is a migration destination. The PC is an edge device on the local network. Host #1 and host #2 are both located at the edge of local network 220 in Fig. 2, and therefore they form an edge computing environment location.)
receiving: i) a description of a training workload comprising a set of virtual machines (VMs) and a set of metadata for different applications to migrate; ([0001], [0005], [0051], [0053]-[0054])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have trained Zolotow’s migration engine using Tan’s workload migration training data to predict a migration destination host . A motivation for the combination is to relieve a source host when it is predicted to be overburdened with computing tasks (Tan, [0001). The combination of Zolotow and Tan would improve Tan’s system because Zolotow’s migration engine would be trained to determine an optimal migration destination host in an end-to-end manner.
However, Zolotow and Tan do not explicitly teach: in response to migrating a workload to the edge computing environment location that is identified based on an output of the trained machine learning algorithm, receiving feedback on performance of the machine learning algorithm by tracking migration metrics in the activity logs at the edge computing environment location.
But Nagpal teaches: in response to migrating a workload to the edge computing environment location that is identified based on an output of the trained machine learning algorithm, receiving feedback on performance of the machine learning algorithm by tracking migration metrics in the activity logs at the edge computing environment location. ([0043], [0053], [0054], lines 1-7, [0055] teaches a provisioning model, which may be a trained neural network, suggests instances for a workload at a second/new computing environment. Performance indicators of a workload executing in a computing environment are collected and used to retrain the provisioning model, where the provisioning model had generated the configuration for the instance of the workload in the computing environment. The feature of “migration metrics in the activity logs” includes performance indicators in the computing environment.)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have applied Nagpal’s feedback loop that retrains a model based on its past predictions to Zolotow/Tan’s model. A motivation for the combination is to improve recommendations produced by a model. (Nagpal, [0055])
Regarding claim 2, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow teaches: wherein the training comprises adjusting a configuration of the machine learning algorithm so as to decrease an [error]
However, Zolotow and Nagpal do not explicitly teach: outage time
But Tan teaches training to decrease an outage time objective ([0007]-[0010] discloses a computing workload on a source host in combination with a migration process exceeds 100% of CPU resources. This results in an outage time because the CPU is unavailable for other tasks. [0105]-[0108] discloses that by implementing proactive workload migration, the CPU resources never exceed 80% and thus the CPU is available for other tasks.)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have incorporated Tan’s outage time objective into the combination of Zolotow, Tan, and Nagpal. A motivation for the combination is the same as the motivation provided for claim 1.
Regarding claim 3, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow teaches: wherein the training comprises: adjusting a configuration of the machine learning algorithm so as to decrease a difference between the identifier of the edge location for migrating the training workload and a second identifier that is output by the machine learning algorithm. ([0043], lines 17-24 discloses adjusting weights and biases of the machine learning module to decrease the error between the output optimal storage environment (“a second identifier that is output by the machine learning component”) and the desired optimal storage environment (“the identifier of the edge location”).)
Regarding claim 4, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow teaches: further comprising computing a business impact of the edge location based on a frequency of access of the workload, ([0055], [0059], from line 11 to “I/O per second” in line 15 and all [0060] discloses determining whether a particular data set for an application is currently stored in an optimal storage environment based on I/O operations per second, where an administrator determines an optimal storage environment for an access pattern type. “Computing a business impact” is performed in step 810.)
wherein the training is based on a loss function for a gradient of the neural network. ([0043], lines 21-26 discloses decreasing an error using gradient descent. In the field of machine learning, a training error is a difference between actual and predicted outputs, which is a loss function.)
Regarding claim 5, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow teaches: wherein the description of the training workload comprises
However, Zolotow and Nagpal do not explicitly teach: a time series of a plurality of variables.
But Tan teaches: the training workload comprises a time series of a [variable]
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have incorporated Tan’s training workload into the combination of Zolotow, Tan, and Nagpal. A motivation for the combination is the same as the motivation provided for claim 1.
Claim 11 recites a product which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
Zolotow teaches: A computer program product for training a machine learning algorithm for identifying an edge computing environment location as a target for workload migration, the computer program product comprising one or more computer readable storage media having program instructions embodied therewith, the program instructions executable by a processing unit to cause the processing unit to perform a method ([0067])
Claims 12-15 each recites a product which implements the same features as the method of claims 2-5, respectively, and are therefore rejected for at least the same reasons.
Claim 21 recites a system which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
Zolotow teaches: A system for training a machine learning algorithm for identifying an edge computing environment location as a target for workload migration, the system comprising: a processor arrangement ([0077], line 3 to “1104” in line 5)
Claims 22-25 each recites a system which implements the same features as the method of claims 2-5, respectively, and are therefore rejected for at least the same reasons.
Regarding claim 26, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow teaches: wherein the training the machine learning algorithm is executed by a computing device that includes software provided as a service in a cloud environment. ([0042], lines 1-7 teaches training. All of [0076] teaches storage server 100 executes program modules and may be implemented in distributed cloud computing environments where tasks are performed by remote processing devices. A “computing device” includes storage server 100 and computer system 1102.)
Regarding claim 29, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1, further comprising:
However, Zolotow does not explicitly teach: analyzing workloads running at the edge computing environment location; and dynamically scheduling, based on the analyzing, migration of the workload to optimize computational resources at the edge computing environment location.
But Tan teaches: analyzing workloads running at the edge computing environment location; and ([0105]-[0109] discloses analyzing a computing workload on the source host and predicting when it will exceed 80% of available computing resources. The source host is at the edge of local network 220 in Fig. 2.)
dynamically scheduling, based on the analyzing, migration of the workload to optimize computational resources at the edge computing environment location. ([0107]-[0109])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have applied Tan’s schedule for migration into the combination of Zolotow, Tan, and Nagpal. A motivation for the combination is the same as the motivation given for claim 1.
Nagpal at [0053] teaches migrating workloads to optimize computing resources.
Claim 32 recites a product which implements the same features as the method of claim 29 and is therefore rejected for at least the same reasons.
Claim 35 recites a system which implements the same features as the method of claim 29 and is therefore rejected for at least the same reasons.
Claims 27, 30, and 33 are rejected under 35 U.S.C. 103 as being unpatentable over Zolotow et al. (US 20200019311 A1, cited in PTO-892 issued 06/20/2025) in view of Tan (US 20100169253 A1, cited in PTO-892 issued 06/20/2025), Nagpal et al. (US 20220156114 A1), and Lan (US 20220138566 A1).
Regarding claim 27, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
Zolotow at [0042]-[0043], [0049] teaches “the received description of the training workload” as input data for training, and “the received identifier of the edge location” as target data for training. However, Zolotow, Tan, and Nagpal do not explicitly teach: further comprising formatting the received description of the training workload and the received identifier of the edge location into a common format that preserves any relationships therebetween.
But Lan teaches: further comprising formatting the [input data and target data]
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have formatted Zolotow’s training data using Lan’s table. A motivation for the combination is to organize training data in storage. (Lan, [0067]-[0068])
Claim 30 recites a product which implements the same features as the method of claim 27 and is therefore rejected for at least the same reasons.
Claim 33 recites a system which implements the same features as the method of claim 27 and is therefore rejected for at least the same reasons.
Claims 28, 31, and 34 are rejected under 35 U.S.C. 103 as being unpatentable over Zolotow et al. (US 20200019311 A1, cited in PTO-892 issued 06/20/2025) in view of Tan (US 20100169253 A1, cited in PTO-892 issued 06/20/2025), Nagpal et al. (US 20220156114 A1), and Chen (“Neural Ordinary Differential Equations”).
Regarding claim 28, the combination of Zolotow, Tan, and Nagpal teaches: The method of claim 1,
However, Zolotow, Tan, and Nagpal do not explicitly teach: wherein the machine learning algorithm is a neural ordinary differential equation that uses a differential equation solver to generate the output of the trained machine learning algorithm.
But Chen teaches: wherein the machine learning algorithm is a neural ordinary differential equation that uses a differential equation solver to generate the output of the trained machine learning algorithm. (Page 1, § 1 to the end of the page)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have applied Chen’s neural ordinary differential equation for the machine learning algorithm in the combination of Zolotow, Tan, and Nagpal. A motivation for the combination is that neural ODEs are memory efficient. (Chen, page 1, § 1, paragraph titled “Memory efficiency”)
Claim 31 recites a product which implements the same features as the method of claim 28 and is therefore rejected for at least the same reasons.
Claim 34 recites a system which implements the same features as the method of claim 28 and is therefore rejected for at least the same reasons.
Response to Arguments
Examiner responds to Applicant’s arguments filed 04/06/2026.
Applicant’s Arguments Under 35 U.S.C. 101: On pages 9-12, Applicant argues the limitations of integrates any judicial exceptions into a practical application.
Examiner’s Response: Applicant’s arguments have been fully considered and are persuasive. The rejection of claim 1 has been withdrawn.
Applicant’s Arguments Under 35 U.S.C. 103: On page 14, Applicant argues the combination of Zolotow and Tan does not teach the steps of claim 1, lines 12-15. The cited portions of Zolotow disclose training a machine learning module configured to receive some combination of the data access information for a data set as input and produce an access pattern classification as output that is provided to a rules engine for processing a set of rules to determine an optimal storage environment based on the access pattern classifications.
Tan fails to cure the deficiencies of Zolotow. In Tan, plot lines 710 and 720 in FIG. 7 of Tan respectively represent computing workloads on a source host where no computing tasks are migrated away therefrom and where a number of computing tasks are proactively migrated away from the source host to a destination host. Therefore, any difference between the plot lines 710 and 720 can only represent a set of computing tasks migrated away from the source host, rather than any set of migration metrics detected in activity logs at the edge computing environment location.
Examiner’s Response: Applicant's arguments have been fully considered but they are not persuasive. The limitations “migrating a workload to the edge computing environment location that is identified based on an output of the trained machine learning algorithm” and “performance of the machine learning algorithm” are taught by Zolotow, [0042], lines 1-7 and [0060], where the output of the trained machine learning module plus rules engine (“the trained machine learning algorithm”) identifies an optimal storage environment, and the system migrates the data set to the optimal storage environment. The task of identifying is a performance.
Applicant’s arguments with respect to the claim 1 limitation “in response to migrating a workload… , receiving feedback on performance of the machine learning algorithm by tracking migration metrics in the activity logs at the edge computing environment location” 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 amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 Asher H. Jablon whose telephone number is (571)270-7648. The examiner can normally be reached Monday - Friday, 9:00 am - 6:00 pm.
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, Abdullah Al Kawsar can be reached at (571)270-3169. 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.
/A.H.J./Examiner, Art Unit 2127
/ABDULLAH AL KAWSAR/Supervisory Patent Examiner, Art Unit 2127
Prosecution Insights