System to infer longevity of cloud computing resource usage and rank in order of importance

Patent metadata

Field	Value
Publication number	US-12505395-B2
Application number	US-202318455435-A
Country	US
Kind code	B2
Filing date	Aug 24, 2023
Priority date	Aug 24, 2023
Publication date	Dec 23, 2025
Grant date	Dec 23, 2025

Abstract

An approach is disclosed that creates a network graph that identifies computing resources as nodes. Each of the computing resources is weighted based upon the corresponding resource's attributes and usage. The approach connects a first set of nodes as directly connected and a second set of nodes as indirectly connected. Node longevity values are calculated for each node in the network graph with each of the node longevity values corresponding to one of the nodes in the network graph. The calculations are based on the direct and indirect connections between each of the nodes. The computing resources are managed based on the corresponding node longevity values.

First claim

What is claimed is:

1 . A method, implemented by a processor coupled to a memory, comprising:
creating a network graph that identifies a plurality of computing resources as nodes, wherein each of the computing resources is weighted based upon the corresponding resource's attributes and usage; connecting first one or more pluralities of nodes as directly connected and a second one or more pluralities of nodes as indirectly connected; calculating a plurality of node longevity values, each of the node longevity values corresponding to one node of the plurality of nodes in the network graph, wherein the calculating is based on the direct and indirect connections between each of the plurality of nodes, wherein the calculating of a node longevity value for the one node of the plurality of nodes is based on factors comprising an age, a read/write usage, a size, and a health score of the one node of the plurality of nodes; and automatically migrating the computer resources in phases in accordance with the plurality of node longevity values such that some of the computer resources are not migrated thereby managing the computing resources based on the corresponding node longevity values, wherein the managing further comprises performing system upgrades, software updates, or security checks based on a set of short-term resources.

2 . The method of claim 1 further comprising:
identifying an amount of longevity influence corresponding to each of the resource attributes and usages, wherein the weight is based on the identified amount of longevity influence.

3 . The method of claim 1 further comprising:
identifying a total number of indirect connections between a first node and a second node, wherein the calculating of the longevity values uses a square root of the total number of indirect connections.

4 . The method of claim 1 , wherein the managing is a management of a cloud migration activity, wherein the method further comprises:
categorizing the plurality of resources based on their corresponding node longevity values; identifying a set of short-term resources based on the categorizing, wherein the set of short-term resources are evaluated based on a next action recommended for each of the short-term resources; identifying a set of low-importance resources based on a usage forecast, wherein the set of low-importance resources are evaluated based on a next action recommended for each of the low-importance resources, wherein at least one of the low-importance resources is recommended to be destroyed; and generating a cloud migration usage forecast based on the resource longevity values, the identified set of short-term resources, and the identified set of low-importance resources.

5 . The method of claim 1 , wherein the managing is a cloud computing recommendation, wherein the method further comprises:
categorizing the plurality of resources based on their corresponding node longevity values; identifying a set of high-importance resources based on the categorizing, and a set of recommendations corresponding resources; identifying a plurality of short-term resources and a plurality of long-term resources based on the resource's respective node longevity values; recommending a cloud computing action corresponding to the plurality of short-term resources; generating a cloud computing forecast based on the resource longevity values, the cloud computing forecast providing recommendations corresponding to the computing resources based on the corresponding resource's longevity value and a prediction on a time to shutdown one or more of the computing resources.

6 . The method of claim 1 , wherein the managing is a resource maintenance recommendation, wherein the method further comprises:
categorizing the plurality of resources based on their corresponding node longevity values; identifying a high value set of resources that are of high-importance based on higher longevity values and based on high usage values, wherein the high value set of resources are scheduled for automated resource maintenance; and identifying a low value set of resources that are of low-importance based on lower longevity values and based on low usage values, wherein the low value set of resources are scheduled for low priority maintenance.

7 . The method of claim 1 , wherein the managing is a maturity assessment, wherein the method further comprises:
summarizing a plurality of datapoints corresponding to the plurality of resources into a plurality of aggregated metrics; inputting the aggregated metrics into an artificial intelligence (Al) model trained to understand maturity assessment based on a financial operations (FinOps) standard; receiving a set of maturity assessment scores from the trained Al model, wherein the maturity assessment scores include a current usage pattern corresponding to the plurality of resources and a future usage pattern corresponding to the plurality of resources; and creating a maturity assessment forecast, wherein the maturity assessment forecast is based on the set of maturity assessment scores.

8 . The method of claim 1 , wherein:
the node longevity value increases as the age increases; the node longevity value increases as the read/write usage increases; the node longevity value increases as the size increases; and the node longevity value increases as the health score increases.

9 . An information handling system comprising:
one or more processors; a memory coupled to at least one of the processors; and a set of instructions stored in the memory and executed by at least one of the processors to perform actions comprising: creating a network graph that identifies a plurality of computing resources as nodes, wherein each of the computing resources is weighted based upon the corresponding resource's attributes and usage; connecting first one or more pluralities of nodes as directly connected and a second one or more pluralities of nodes as indirectly connected; calculating a plurality of node longevity values, each of the node longevity values corresponding to one node of the plurality of nodes in the network graph, wherein the calculating is based on the direct and indirect connections between each of the plurality of nodes, wherein the calculating of a node longevity value for the one node of the plurality of nodes is based on factors comprising an age, a read/write usage, a size, and a health score of the one node of the plurality of nodes; and automatically migrating the computer resources in phases in accordance with the plurality of node longevity values such that some of the computer resources are not migrated thereby managing the computing resources based on the corresponding node longevity values, wherein the managing further comprises performing system upgrades, software updates, or security checks based on a set of short-term resources.

10 . The information handling system of claim 9 wherein the actions further comprise:
identifying an amount of longevity influence corresponding to each of the resource attributes and usages, wherein the weight is based on the identified amount of longevity influence.

11 . The information handling system of claim 9 wherein the actions further comprise:
identifying a total number of indirect connections between a first node and a second node, wherein the calculating of the longevity values uses a square root of the total number of indirect connections.

12 . The information handling system of claim 9 , wherein the managing is a management of a cloud migration activity, wherein the actions further comprise:
categorizing the plurality of resources based on their corresponding nod

Assignees

• Kyndryl Inc

Inventors

• Wheatley Keri
• Chintakunta Mouleswara Reddy
• Odibat Omar

Classifications

• G06Q10/06315 Primary

  Needs-based resource requirements planning or analysis · CPC title