Methods and system for identifying a customer profile based on metadata

A method for managing a data protection module, the method comprising:

• obtaining, by an analyzer, client metadata associated with a client environment (CE) data protection module by making periodic application programming interface (API) calls to the CE data protection module, wherein the client metadata specifies a type of a workload utilized by a user, computing resource utilization data regarding resources of the CE data protection module, data protection policies implemented by the user, and deployment configurations set by the user for the CE data protection module;
• obtaining, by the analyzer, vendor metadata associated with a vendor environment (VE) data protection module by making periodic API calls to the VE data protection module, wherein the vendor metadata specifies a type of a workload tested by an administrator of the VE data protection module, a recommended resource utilization value by a vendor, data protection policies tested on the VE data protection module, and a deployment environment tested on the VE data protection module;
• analyzing, by the analyzer, the client metadata and the vendor metadata to extract relevant data using a group of linear and non-linear machine learning models, wherein one or more of the group of linear and non-linear machine learning models is a set difference model, wherein the group of linear and non-linear machine learning models are retrained periodically using updated training data;
• making, by the analyzer and based on the relevant data, a first determination that a mismatch exists between the client metadata and the vendor metadata;
• making, by the analyzer and based on the first determination, a second determination that the mismatch is a client-related mismatch;
• obtaining, by the analyzer, using the group of linear and non-linear machine learning models, and based on the client-related mismatch, a first similarity score for a first resource and a second similarity score for a second resource, wherein the client metadata specifies at least the first resource and the second resource, and wherein one or more of the group of linear and non-linear machine learning models is a node similarity model;
• identifying, by the analyzer and based on the first similarity score and the second similarity score, that the first similarity score is higher than the second similarity score;
• sending, by the analyzer and based on the identification, a recommendation to the administrator of the VE data protection module to manage the client-related mismatch via a graphical user interface of the VE data protection module;
• and
• monitoring, via an operation monitoring service, the CE data protection module using an application programming interface to determine whether the recommendation resolved the client-related mismatch, wherein the operation monitoring service is a computer program executing on underlying hardware of the analyzer.

A system for managing a data protection module, the system comprising:

• a processor comprising circuitry;
• memory comprising instructions, which when executed perform a method, the method comprising:
• obtaining, by an analyzer, client metadata associated with a client environment (CE) data protection module by making periodic application programming interface (API) calls to the CE data protection module, wherein the client metadata specifies a type of a workload utilized by a user, computing resource utilization data regarding resources of the CE data protection module, data protection policies implemented by the user, and deployment configurations set by the user for the CE data protection module;
• obtaining, by the analyzer, vendor metadata associated with a vendor environment (VE) data protection module by making periodic API calls to the VE data protection module, wherein the vendor metadata specifies a type of a workload tested by an administrator of the VE data protection module, a recommended resource utilization value by a vendor, data protection policies tested on the VE data protection module, and a deployment environment tested on the VE data protection module;
• analyzing, by the analyzer, the client metadata and the vendor metadata to extract relevant data using a group of linear and non-linear machine learning models, wherein one or more of the group of linear and non-linear machine learning models is a set difference model, wherein the group of linear and non-linear machine learning models are retrained periodically using updated training data;
• making, based on the relevant data, a determination that a mismatch exists between the client metadata and the vendor metadata;
• obtaining, by the analyzer, using the group of linear and non-linear machine learning models, and based on the determination, a first similarity score for a first resource and a second similarity score for a second resource, wherein the client metadata also specifies at least the first resource and the second resource, and wherein one or more of the group of linear and non-linear machine learning models is a node similarity model;
• identifying, by the analyzer and based on the first similarity score and the second similarity score, that the first similarity score is lower than the second similarity score;
• sending, based on the identification, a recommendation to manage the mismatch to at least one of a user of the VE data protection module and a user of the CE data protection module;
• and
• monitoring, via an operation monitoring service, the CE data protection module using an application programming interface to determine whether the recommendation resolved the mismatch, wherein the operation monitoring service is a computer program executing on underlying hardware of the analyzer.