# US 2026/0093588 A1 · Intelligent Profile-Driven Drift Detection > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2026/0093588 A1 - Document ID: US20260093588A1 - Status: Published application - Kind: A1 - Application number: 18/901,506 - Filed: 2024-09-30 - Published: 2026-04-02 - Earliest claimed: 2024-09-30 - Assignees: Oracle International Corporation - Applicants: Oracle International Corporation - Inventors: Gaurav Kumar; Prashant Kumar; Nagarajan Muthukrishnan; Prasanna Venkatesh Ramamurthi; Binoy Sukumaran; Nirmala Sreekantaiah - Web page: https://patentgrep.com/patent/US20260093588A1 ## CPC - G06F 11/3051: {Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs (verification or detection of system hardware configuration G06F11/2247)} - G06F 11/3006: {where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50)} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Techniques are disclosed for detecting a drift experienced by computing system(s). The system generates multiple snapshots as part of a drift detection process. Each snapshot contains state information of a computing system. Based on the snapshots, the system generates metrics sets according to a general specification. The general specification defines metrics generally suitable for detecting drift in the computing system(s). Based on the circumstances of the drift detection process, the system generates a custom specification. The system optionally employs trained machine learning model(s) for custom specification generation. The custom specification defines modifications to the metric sets designed to make the metric sets more suitable for the circumstances of the drift detection process. The system modifies the metric sets according to the custom specification. Subsequently, the system generates flattened vectors based on the modified metric sets, and the system performs a cluster analysis on the flattened vectors to detect any drift. ## Claims ### Claim 1 One or more non-transitory computer-readable media comprising instructions that, when executed by one or more hardware processors, cause performance of operations comprising: generating a first plurality of snapshots comprising state information of a set of target systems, the set of target systems comprising one or more target systems, wherein the first plurality of snapshots comprises at least one of: (a) a second plurality of snapshots that represents a respective plurality of target systems or (b) a third plurality of snapshots that represents a particular target system at a respective plurality of points in time; based, at least in part, on the first plurality of snapshots, generating a plurality of metric sets according to a first specification, wherein the first specification defines a plurality of metrics for drift detection; based, at least in part, on a feature set, modifying the plurality of metric sets to generate a modified plurality of metric sets, wherein modifying the plurality of metric sets comprises at least one of: (a) adding weight to a first metric comprised within a first metric set, (b) removing weight from a second metric comprised within the first metric set, (c) adding a third metric to the first metric set, or (d) removing a fourth metric from the first metric set; and based, at least in part, on the modified plurality of metric sets, detecting a drift associated with the set of target systems. ### Claim 2 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein the set of target systems comprises a first target system that corresponds to the first metric set; wherein the drift impacts the first target system; wherein the first target system comprises a first hierarchy of resources; wherein the first metric set indicates a first state of the first hierarchy of resources; and wherein the drift is characterized by at least one of (a) an abnormally configured resource comprised within the first hierarchy of resources, (b) an abnormally allocated resource comprised within the first hierarchy of resources, or (c) an abnormally utilized resource comprised within the first hierarchy of resources. ### Claim 3 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein generating the modified plurality of metric sets comprises: determining a first dataset based on a first snapshot of a first target system, the first dataset comprising sensitive information associated with the first target system, wherein removing the sensitive information from a first computing environment comprising the first target system is prohibited; applying one or more hashing algorithms to the first dataset to generate a first hash, wherein the first metric set (a) represents the first target system and (b) comprises the first hash; aggregating the plurality of metric sets, the plurality of metric sets comprising the first metric set, wherein aggregating the plurality of metric sets comprises transmitting the first metric set from the first computing environment to a second computing environment. ### Claim 4 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein detecting the drift associated with the set of target systems comprises: normalizing values comprised within the plurality of metric sets; generating a plurality of vectors based, at least in part, on the values comprised within the plurality of metric sets; applying one or more clustering algorithms to the plurality of vectors to generate a set of clusters, the set of clusters comprising one or more clusters; and analyzing the set of clusters to detect the drift associated with the set of target systems. ### Claim 5 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein the feature set comprises at least one of: (a) an attribute of a target system; (b) a first user characteristic of a first user associated with the target system; (c) a second user characteristic of a second user that initiates an ongoing drift detection process; (d) a first set of historical data pertaining to the set of target systems; (d) a second set of historical data pertaining to one or more prior drift detection processes; (e) a type of drift potentially associated with the set of target systems; (f) user input; (g) a feature derived from applying natural language processing to a natural language input; or (f) information comprised within the plurality of metric sets. ### Claim 6 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein the operations further comprise: prior to modifying the plurality of metric sets: based, at least in part, on the feature set, generating a second specification, the second specification defining one or more modifications to the plurality of metric sets, wherein generating the second specification comprises: applying a machine learning model to the feature set to select one or more metrics that are relevant to detecting one or more drifts that are potentially associated with the set of target systems, wherein the one or more modifications defined by the second specification are determined based, at least in part, on the machine learning model selecting the one or more metrics; wherein the plurality of metric sets is modified according to the second specification. ### Claim 7 (depends on 6) The one or more non-transitory computer-readable media of claim 6, wherein the operations further comprise: prior to generating the second specification: accessing training data, the training data comprising a first set of training data, wherein the first set of training data defines an association between (a) a first feature of a first drift detection process and (b) a first set of one or more metrics that are relevant to detecting drift if the first feature exists; training the machine learning model to select metrics relevant to drift detection based, at least in part, on the training data; subsequent to generating the second specification: accessing feedback pertaining to the second specification; and further training the machine learning model based, at least in part, on the feedback. ### Claim 8 (depends on 1) The one or more non-transitory computer-readable media of claim 1, wherein subsequent to detecting the drift the operations further comprise at least one of: (a) performing a fine-grain drift analysis involving a subset of target systems comprised within the set of target systems, the subset of target systems comprising a first target system and a second target system, wherein the first target system is impacted by the drift and the second target system conforms to a target state; (b) alerting a user to the drift by presenting a communication to the user, the communication comprising at least one of: (a) an identification of the first target system, (b) a cause of the drift, or (c) one or more suggested operations for remediating the drift; or (c) converting the first target system from a present state to at least one of: (a) a previous state of the first target system or (b) the target state. ### Claim 9 (depends on 2) The one or more non-transitory computer-readable media of claim 2, wherein the first target system is a cloud computing system that is cooperatively managed by a plurality of entities, wherein the plurality of entities comprises a cloud user and a cloud provider; wherein the cloud user manages a first subset of higher-level resources comprised within the first hierarchy of resources; wherein the cloud provider manages a second subset of lower-level resources comprised within the first hierarchy of resources; wherein the first metric set represents both (a) quantitative attributes of the first target system and (b) qualitative attributes of the first target system; wherein the quantitative attributes of the first target system are represented by numerical values comprised within the first metric set; and wherein the qualitative attributes of the first target system are represented by numerical digital signatures comprised within the first metric set. ### Claim 10 A method comprising: generating a first plurality of snapshots comprising state information of a set of target systems, the set of target systems comprising one or more target systems, wherein the first plurality of snapshots comprises at least one of: (a) a second plurality of snapshots that represents a respective plurality of target systems or (b) a third plurality of snapshots that represents a particular target system at a respective plurality of points in time; based, at least in part, on the first plurality of snapshots, generating a plurality of metric sets according to a first specification, wherein the first specification defines a plurality of metrics for drift detection based, at least in part, on a feature set, modifying the plurality of metric sets to generate a modified plurality of metric sets, wherein modifying the plurality of metric sets comprises at least one of: (a) adding weight to a first metric comprised within a first metric set, (b) removing weight from a second metric comprised within the first metric set, (c) adding a third metric to the first metric set, or (d) removing a fourth metric from the first metric set; and based, at least in part, on the modified plurality of metric sets, detecting a drift associated with the set of target systems, wherein the method is performed by at least one device including a hardware processor. ### Claim 11 (depends on 10) The method of claim 10, wherein the set of target systems comprises a first target system that corresponds to the first metric set; wherein the drift impacts the first target system; wherein the first target system comprises a first hierarchy of resources; wherein the first metric set indicates a first state of the first hierarchy of resources; and wherein the drift is characterized by at least one of (a) an abnormally configured resource comprised within the first hierarchy of resources, (b) an abnormally allocated resource comprised within the first hierarchy of resources, or (c) an abnormally utilized resource comprised within the first hierarchy of resources. ### Claim 12 (depends on 10) The method of claim 10, wherein generating the modified plurality of metric sets comprises: determining a first dataset based on a first snapshot of a first target system, the first dataset comprising sensitive information associated with the first target system, wherein removing the sensitive information from a first computing environment comprising the first target system is prohibited; applying one or more hashing algorithms to the first dataset to generate a first hash, wherein the first metric set (a) represents the first target system and (b) comprises the first hash; aggregating the plurality of metric sets, the plurality of metric sets comprising the first metric set, wherein aggregating the plurality of metric sets comprises transmitting the first metric set from the first computing environment to a second computing environment. ### Claim 13 (depends on 10) The method of claim 10, wherein detecting the drift associated with the set of target systems comprises: normalizing values comprised within the plurality of metric sets; generating a plurality of vectors based, at least in part, on the values comprised within the plurality of metric sets; applying one or more clustering algorithms to the plurality of vectors to generate a set of clusters, the set of clusters comprising one or more clusters; and analyzing the set of clusters to detect the drift associated with the set of target systems. ### Claim 14 (depends on 10) The method of claim 10, wherein the feature set comprises at least one of: (a) an attribute of a target system; (b) a first user characteristic of a first user associated with the target system; (c) a second user characteristic of a second user that initiates an ongoing drift detection process; (d) a first set of historical data pertaining to the set of target systems; (d) a second set of historical data pertaining to one or more prior drift detection processes; (e) a type of drift potentially associated with the set of target systems; (f) user input; (g) a feature derived from applying natural language processing to a natural language input; or (f) information comprised within the plurality of metric sets. ### Claim 15 (depends on 10) The method of claim 10, further comprising: prior to modifying the plurality of metric sets: based, at least in part, on the feature set, generating a second specification, the second specification defining one or more modifications to the plurality of metric sets, wherein generating the second specification comprises: applying a machine learning model to the feature set to select one or more metrics that are relevant to detecting one or more drifts that are potentially associated with the set of target systems, wherein the one or more modifications defined by the second specification are determined based, at least in part, on the machine learning model selecting the one or more metrics; wherein the plurality of metric sets is modified according to the second specification. ### Claim 16 (depends on 15) The method of claim 15, further comprising: prior to generating the second specification: accessing training data, the training data comprising a first set of training data, wherein the first set of training data defines an association between (a) a first feature of a first drift detection process and (b) a first set of one or more metrics that are relevant to detecting drift if the first feature exists; training the machine learning model to select metrics relevant to drift detection based, at least in part, on the training data; subsequent to generating the second specification: accessing feedback pertaining to the second specification; and further training the machine learning model based, at least in part, on the feedback. ### Claim 17 (depends on 10) The method of claim 10, further comprising: subsequent to detecting the drift, performing at least one of: (a) performing a fine-grain drift analysis involving a subset of target systems comprised within the set of target systems, the subset of target systems comprising a first target system and a second target system, wherein the first target system is impacted by the drift and the second target system conforms to a target state; (b) alerting a user to the drift by presenting a communication to the user, the communication comprising at least one of: (a) an identification of the first target system, (b) a cause of the drift, or (c) one or more suggested operations for remediating the drift; or (c) converting the first target system from a present state to at least one of: (a) a previous state of the first target system or (b) the target state. ### Claim 18 (depends on 11) The method of claim 11, wherein the first target system is a cloud computing system that is cooperatively managed by a plurality of entities, wherein the plurality of entities comprises a cloud user and a cloud provider; wherein the cloud user manages a first subset of higher-level resources comprised within the first hierarchy of resources; wherein the cloud provider manages a second subset of lower-level resources comprised within the first hierarchy of resources; wherein the first metric set represents both (a) quantitative attributes of the first target system and (b) qualitative attributes of the first target system; wherein the quantitative attributes of the first target system are represented by numerical values comprised within the first metric set; and wherein the qualitative attributes of the first target system are represented by numerical digital signatures comprised within the first metric set. ### Claim 19 A system comprising: at least one device including a hardware processor; the system being configured to perform operations comprising: generating a first plurality of snapshots comprising state information of a set of target systems, the set of target systems comprising one or more target systems, wherein the first plurality of snapshots comprises at least one of: (a) a second plurality of snapshots that represents a respective plurality of target systems or (b) a third plurality of snapshots that represents a particular target system at a respective plurality of points in time; based, at least in part, on the first plurality of snapshots, generating a plurality of metric sets according to a first specification, wherein the first specification defines a plurality of metrics for drift detection based, at least in part, on a feature set, modifying the plurality of metric sets to generate a modified plurality of metric sets, wherein modifying the plurality of metric sets comprises at least one of: (a) adding weight to a first metric comprised within a first metric set, (b) removing weight from a second metric comprised within the first metric set, (c) adding a third metric to the first metric set, or (d) removing a fourth metric from the first metric set; and based, at least in part, on the modified plurality of metric sets, detecting a drift associated with the set of target systems. ### Claim 20 (depends on 19) The system of claim 19, wherein the set of target systems comprises a first target system that corresponds to the first metric set; wherein the drift impacts the first target system; wherein the first target system comprises a first hierarchy of resources; wherein the first metric set indicates a first state of the first hierarchy of resources; and wherein the drift is characterized by at least one of (a) an abnormally configured resource comprised within the first hierarchy of resources, (b) an abnormally allocated resource comprised within the first hierarchy of resources, or (c) an abnormally utilized resource comprised within the first hierarchy of resources.