DETAILED ACTION
This action is in response to the filing 03/12/2026. Claims 1-20 are pending and have been fully examined.
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 2 and 12 have been cancelled.
Claims 3-4, and 7 contain allowable subject matter and are objected to for their dependence upon rejected base claims
Claims 11 and 13-20 are allowed.
Claims 1, 5-6, and 8-10 are rejected under 35 U.S.C. 103.
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, 5-6, and 8-10 are rejected under 35 U.S.C. 103 as being unpatentable over Ambaljeri et al. (U.S. PGPub No. 20240394149) in view of Chopra et al. (U.S. PGPub No. 20240303359).
Regarding Claim 1, Ambaljeri teaches,
A system, comprising: a non-transitory memory having instructions stored thereon; and at least one processor operatively coupled to the non-transitory memory, and configured to read the instructions to: … generate training data for the at least one machine learning model based on… and metadata related to client applications, clusters, topics of the data service platform (metadata obtained and associated with service requests includes: application metadata ("metadata related to client applications") [0191], a type of a workload ("metadata related to clusters [and] topics of the data service platform") [0191]),
train the at least one machine learning model to learn to detect workload problem patterns based on the training data (the models may be retrained periodically to implement improvements to the models [0161]; the examiner notes that if the models are re-trained, the models are previously initially trained; where the models are applied to infer states or the occurrence of states based on pattern information ("workload problem patterns") [0155&0196]);
receive, over a network, an observability request for at least one data service platform that stores messages coming from producer applications (where the analyzer receives a metadata analysis request ("observability request") to initiate analyzing obtained metadata [0195]; where the analyzer and client device communicate over the network [0022])
and, in response, receive metadata for the at least one data service platform that includes data related to at least two of: one or more clusters in the workload; one or more topics hosted on the one or more clusters; a number of partitions for each of the one or more topics; one or more consumer applications consuming each of the one or more topics; and one or more configurations of the one or more clusters and the one or more topics (configuration parameters ("configurations of the one or more clusters and the topics") [0141], partition assignment strategy [0152], and application type including active sessions ("a list of consumer applications consuming each topic") [0169]; the examiner notes that the above list includes three items ("at least two of:..."));
determine, based on a catalog of problem patterns and the metadata of the workload, that a problem pattern exists in the workload using the trained at least one machine learning model (using machine learning models applied to obtained metadata, inferring previous or current states and that events have occurred [0155]; pattern information is generated based on obtained metadata [0196]);
generate observability data that identifies a problem instance for the workload in accordance with the determination that the problem pattern exists in the workload (pattern information is generated and used to identify failure information, including remediation [0196-0197]);
create a problem record for the problem instance based on the observability data; store the problem record in a database (the pattern database is updated with the new record, including resolution and result (problem record) [0201]);
and, transmit, over the network, the problem record in response to the observability request (the generated recommendation (resolution; "problem record") is received over the network [0155]; the examiner notes that therefore, the recommendation is transmitted over the network).
Ambaljeri does not appear to disclose and Chopra teaches,
generate training data for at least one machine learning model based on historical problem data, historical detected problem instances, historical solution data, historical or labelled problem data and solution data, ... (where the models may be retrained using "any form of training data" [0164]; training data may include: previously obtained application and system logs ("historical detected problem instances/data") [0206] and previously obtained alerts ("labelled problem instances") [0208]; where the database comprises mappings between received alerts and the corresponding recommended solution [0148] and updating-training data is collected from previously stored database [0206&0208]);
It would have been obvious to one of ordinary skill, before the effective filing date of the claimed invention, to modify the problem pattern detection of Ambaljeri with the machine learning model pattern detection training of Chopra. The resulting combination allows for the pattern-analyzing models to be updated as improvements are available [Chopra; 206 & 208], where the pattern-analyzing models allow for an improved user experience with the vendor [0220].
Regarding Claim 5, Ambaljeri does not appear to disclose and Chopra teaches,
The system of claim 1, wherein determining whether a problem pattern exists in the workload comprises: analyzing the metadata using a plurality of problem pattern rules each associated with a corresponding problem pattern in the catalog of problem patterns (the metadata is analyzed with a machine learning model according to set policy/parameters ("rules") to, at least, infer failures and reasons for failure [0204]);
and determining whether a problem pattern exists in the workload based on a corresponding problem pattern rule (the analyzed metadata is used to determine a failure prompting an alert, including type of alert (the examiner notes, therefore type of problem, or specific problem pattern) [0207]).
It would have been obvious to one of ordinary skill, before the effective filing date of the claimed invention, to modify the problem pattern detection of Ambaljeri with the machine learning model pattern detection of Chopra. The resulting combination allows for the pattern-analyzing models to improve user experience with the vendor [0220].
Regarding Claim 6, Ambaljeri teaches,
The system of claim 5, wherein: the analyzing is executed based on at least one of: a periodic configuration, a consumer alert, or a user request; and the metadata comprises relevant metrics from observability data sources of the at least one data service platform (analyzing may be prompted by, at least: a service request ("consumer alert") [0195; also see 0067], a fixed-time interval metadata "dump" to the analyzer ("periodic configuration") [0193; also see 0194], and a metadata analysis request ("user request") [0194[; where metadata comprises relevant metrics [0191]).
Regarding Claim 8, Ambaljeri does not appear to disclose and Chopra teaches,
The system of claim 1, wherein: the at least one machine learning model is trained based on: a predetermined set of problem patterns, historical detected problem instances and/or labelled problem instances (the analyzer (comprising the machine learning models [0162]) may be trained using any form of training data, including: previously obtained application and system logs ("historical detected problem instances") [0206] and previously obtained alerts ("labelled problem instances") [0208]).
The same motivation for Claim 1 also applies to Claim 8.
Regarding Claim 9, Ambaljeri teaches,
The system of claim 1, wherein: the workload is monitored during either a development stage or a production stage of the at least one data service platform (the system may be applied to production workloads [0073]).
Regarding Claim 10, Ambaljeri teaches,
The system of claim 1, wherein the at least one processor is configured to: present the problem instance to a user via an application programming interface (API) (the system components may communicate via APIs [0057], where the system components of Fig. 1 include CE storage [0034]; where the CE storage sends the notification to display the problem notification and resolution information/requests to the user [0080] as well as device status [185]);
determine a problem solution based on the problem instance and a catalog of problem solutions, wherein the problem solution is associated with the problem pattern existing in the workload (the determined pattern information is compared to the service request pattern database ("catalog of problem solutions") [0137]);
and execute the problem solution to recover the workload (where the system may implement determined problem solutions [0201]; the examiner notes, "without the required use of the administrator," or automatically).
Allowable Subject Matter
Claims 3-4 and 7 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The following is the Examiner’s statement for indicating allowable subject matter:
Regarding Claim 3, Ambaljeri discloses configuration parameters ("configurations of the one or more clusters and the topics") [0141], partition assignment strategy [0152], and application type including active sessions ("a list of consumer applications consuming each topic") [0169]. The prior art of record does not disclose, without the use of impermissible hindsight reasoning, the metadata of the workload comprises data related to: one or more clusters in the workload; a list of topics hosted on the one or more clusters … .
Regarding Claim 4, Ambaljeri discloses use of pattern information based on metadata [0196] as applied with a machine learning model to obtain previous or current states and events that have occurred [0155]. In addition to the previously cited prior art of record, the Examiner also points to the following relevant prior art: Vuda (U.S. PGPub No. 20240291718) as teaching an anomaly detection criteria includes threshold values for individual characteristics of nodes and edges of a network, where the characteristics exemplary include the load at a node being greater than a threshold and in response, workloads may be modified, deleted, reconfigured, and/or scaled [0079-0080]. However, the prior art of record does not disclose, without the use of impermissible hindsight reasoning [emphasis added], the catalog of problem patterns includes a juggler pattern where a workload is deployed with a less number of consumer instances across consumer applications than a total number of partitions from which messages are to be consumed, a time slicer pattern where a workload is deployed with consumer applications provisioned with a less number of processor cores than a number of consumer instances configured per consumer application a headline pattern where a same topic in a workload is consumed by multiple consumer applications more than a predetermined threshold, …, a quiescent topic pattern where a topic is not consumed by any consumer application for a time period longer than a predetermined threshold and a diehard client pattern where a client application is implemented using an unsupported version of client library.
Regarding Claim 7, Chopra discloses where service level agreements are used to specify performance requirements, including latency requirements [0038]; where collected data may include: I/O data per second ("incoming messages per second") and computing resource utilization ("processor utilization") [0073]. The prior art of record does not disclose, without the use of impermissible hindsight reasoning, the analyzing is executed based on an alert of consumer lag … and the analyzing comprises analyzing the relevant metrics to determine whether there is a variation in trends of the relevant metrics before and after the consumer lag.
Claims 11 and 13-20 are allowed. The following is the Examiner’s statement of reasons for indicating allowable subject matter:
Regarding Independent Claims 11 and 20, Ambaljeri discloses metadata obtained and associated with service requests includes: application metadata ("metadata related to client applications") [0191], a type of a workload ("metadata related to clusters [and] topics of the data service platform") [0191]; the models may be retrained periodically to implement improvements to the models [0161]; the examiner notes that if the models are re-trained, the models are previously initially trained; where the models are applied to infer states or the occurrence of states based on pattern information ("workload problem patterns") [0155&0196]; where the analyzer receives a metadata analysis request ("observability request") to initiate analyzing obtained metadata [0195]; where the analyzer and client device communicate over the network [0022]; configuration parameters ("confgurations of the one or more clusters and the topics") [0141], partition assignment strategy [0152], and application type including active sessions ("a list of consumer applications consuming each topic") [0169]; using machine learning models applied to obtained metadata, inferring previous or current states and that events have occurred [0155]; pattern information is generated based on obtained metadata [0196]; pattern information is generated and used to identify failure information, including remediation [0196-0197]; the pattern database is updated with the new record, including resolution and result (problem record) [0201] the generated recommendation (resolution; "problem record") is received over the network [0155]; the examiner notes that therefore, the recommendation is transmitted over the network. Chopra discloses where the models may be retrained using "any form of training data" [0164]; training data may include: previously obtained application and system logs ("historical detected problem instances/data") [0206] and previously obtained alerts ("labelled problem instances") [0208]; where the database comprises mappings between received alerts and the corresponding recommended solution [0148] and updating-training data is collected from previously stored database [0206&0208]. The prior art of record does not disclose alone or in combination, without the use of impermissible hindsight reasoning, all elements of the inclusive metadata list, specifically: metadata for the at least one data service platform that includes data related to at least two of: one or more clusters in the workload; one or more topics hosted on the one or more clusters … . Claims 13-19 depend upon the allowable subject matter of Claim 11.
Response to Arguments
Applicant’s remarks filed 03/12/2026 have been fully considered.
The Applicant argues that subject matter previously allowable over prior art (from Claims 3, and similarly, Claim 13) have been amended into independent Claims 1, 11, and 20. The Examiner agrees that the allowable subject matter of Claims 3/13 have been amended into independent Claims 11 and 20 and respectfully points to the Allowable Subject Matter section above. However, the Examiner finds that the inclusive list of Claim 3 has not been amended into Claim 1 with sufficient similarity and instead recites, with emphasis added in italics, metadata for the at least one data service platform that includes data related to at least two of […]. The Examiner finds that Ambaljeri teaches three items of the metadata list, please see the rejection under 35 U.S.C. 103 above. Accordingly, the Examiner respectfully disagrees that the subject matter of Claim 1 and rejected dependent claims are allowable and maintains the art rejection.
Regarding the previous rejection under 35 U.S.C. 101, the Applicant argues at least that the claims recite a technological solution to a technological problem. The Examiner finds that [0003-0004] of the specification describe the technological problem of a lack of cohesive monitoring of large-scale distributed systems where best practices are not followed in deployed workloads, where existing methods to identify and resolve problems are both time-consuming and prone to errors. The Examiner notes that the judicial exception itself (determining... that a problem exists in the workload) cannot provide the improvement alone, see 2106.05(a). However, the Examiner finds that the combination of receiv[ing] metadata for the at least one data service platform …; determin[ing …] based on … the metadata … that a problem pattern exists in the workload using the trained machine learning model; generat[ing] observability data … in accordance with the determination that the problem pattern exists …; creat[ing] a problem record … based on the observability data; and transmit[ting] … the problem record in response to the observability request integrate the judicial exception into practical application by reciting the technical solution as made obvious to one of ordinary skill in the art from [0005, 0028-0029] of the specification; the Examiner finds that the above limitations sufficiently reflect a cohesive monitoring that automatically identifying problem instances in data service workloads to prevent incidents and events leading into incidents. Accordingly, the rejection under 35 U.S.C. 101 is withdrawn.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AUDREY E WHITESELL whose telephone number is (703)756-4767. The examiner can normally be reached 8:30am - 5:00pm MST.
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, Bryce Bonzo can be reached at 5712723655. 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.E.W./Examiner, Art Unit 2113 /BRYCE P BONZO/Supervisory Patent Examiner, Art Unit 2113