US-12619921-B2 - Predictive fog data center migration

Abstract

A computer-implemented method includes: obtaining, by a computing device, data items from data sources; classifying, by the computing device, the data items into categories using a first machine learning (ML) model; generating, by the computing device, a risk score of a first data center based on the classified data items and using a second machine learning (ML) model; determining, by the computing device, the risk score of the first data center exceeds a threshold; and in response to the determining the risk score of the first data center exceeds the threshold, initiating, by the computing device, a migration of the first data center to a second data center.

Inventors

• Girish M Chawla
• Shikhar KWATRA
• Krishna Teja Rekapalli
• Isac Silva
• Russell C. Norberg
• Teresa M. Taylor

Assignees

• KYNDRYL, INC.

Dates

Publication Date

20260505

Application Date

20220830

Claims (20)

1. 1 . A method, comprising: obtaining, by a computing device, data items from data sources; configuring, by the computing device, a weightage of the data items based on a geographic location of the data sources; classifying, by the computing device, the data items into categories using a first machine learning (ML) model; generating, by the computing device, a risk score of a first fog data center based on the classified data items and using a second ML model, wherein normalized respective numbers of the classified data items are used as inputs to the second ML model; determining, by the computing device, the risk score of the first fog data center exceeds a threshold; in response to the determining the risk score of the first fog data center exceeds the threshold, sending to a user, by the computing device, an alert indicating the risk score of the first fog data center and having a selectable field for approving a migration of the first fog data center to a second fog data center; and in response to the user selecting the selectable field approving the migration, initiating, by the computing device, the migration of the first fog data center to the second fog data center before an event occurs, wherein the event comprises one of: failures within the first fog data center and disasters affecting the first fog data center.
2. 2 . The method of claim 1 , further comprising determining respective numbers of the classified data items in respective ones of the categories.
3. 3 . The method of claim 2 , further comprising normalizing the respective numbers of the classified data items.
4. 4 . The method of claim 1 , wherein the first ML model comprises a naïve bayes classifier.
5. 5 . The method of claim 1 , wherein the second ML model comprises a logistic regression model.
6. 6 . The method of claim 1 , further comprising: training a naïve bayes classifier using first training data to create the first ML model; and training a logistic regression model using second training data to create the second ML model, the second training data different from the first training data.
7. 7 . The method of claim 1 , further comprising filtering the data items based on a predefined amount of time.
8. 8 . The method of claim 1 , further comprising filtering the data items based on geographic location relative to the first fog data center.
9. 9 . The method of claim 1 , further comprising determining the geographic location of the data sources by one of: extracting machine readable geographic metadata associated with each data item of each data source; and extracting geographic information using natural language processing from each data item of each data source.
10. 10 . The method of claim 1 , wherein the initiating the migration of the first fog data center to the second fog data center is based on a second risk score of the second fog data center being less than the threshold.
11. 11 . The method of claim 1 , wherein the initiating the migration comprises automatically starting the migration.
12. 12 . The method of claim 1 , wherein the data items comprise weather data and news data.
13. 13 . A computer program product comprising one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to: obtain data items from data sources; configure a weightage of the data items based on a geographic location of the data sources; classify the data items into categories using a first machine learning (ML) model; generate a risk score of a first fog data center based on the classified data items and using a second ML model, wherein normalized respective numbers of the classified data items are used as inputs to the second ML model; determine the risk score of the first fog data center exceeds a threshold; in response to the determining the risk score of the first fog data center exceeds the threshold, send to a user, an alert indicating the risk score of the first fog data center and having a selectable field for approving a migration of the first fog data center to a second fog data center; and in response to the user selecting the selectable field approving the migration, initiate the migration of the first fog data center to the second fog data center before an event occurs, wherein the event comprises one of: failures within the first fog data center and disasters affecting the first fog data center.
14. 14 . The computer program product of claim 13 , wherein the program instructions are executable to: determine respective numbers of the classified data items in respective ones of the categories; and normalize the respective numbers of the classified data items.
15. 15 . The computer program product of claim 13 , wherein: the first ML model comprises a naïve bayes classifier; and the second ML model comprises a logistic regression model.
16. 16 . The computer program product of claim 13 , wherein the program instructions are executable to: filter the data items based on a predefined amount of time; and filter the data items based on geographic location relative to the first fog data center.
17. 17 . A system comprising: a processor, a computer readable memory, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to: obtain data items from data sources; configure a weightage of the data items based on a geographic location of the data sources; classify the data items into categories using a first machine learning (ML) model; generate a risk score of a first fog data center based on the classified data items and using a second ML model, wherein normalized respective numbers of the classified data items are used as inputs to the second ML model; determine the risk score of the first fog data center exceeds a threshold; in response to the determining the risk score of the first fog data center exceeds the threshold, send to a user, an alert indicating the risk score of the first fog data center and having a selectable field for approving a migration of the first fog data center to a second fog data center; and in response to the user selecting the selectable field approving the migration, initiate the migration of the first fog data center to the second fog data center before an event occurs, wherein the event comprises one of: failures within the first fog data center and disasters affecting the first fog data center.
18. 18 . The system of claim 17 , wherein the first ML model comprising a naïve bayes classifier; and the second ML model comprising a logistic regression model.
19. 19 . The system of claim 17 , wherein the program instructions are executable to: filter the data items based on a predefined amount of time; and filter the data items based on geographic location relative to the first fog data center.
20. 20 . The system of claim 17 , wherein the program instructions are executable to: determine respective numbers of the classified data items in respective ones of the categories; and normalize the respective numbers of the classified data.

Description

BACKGROUND Aspects of the present invention relate generally to managing data centers and, more particularly, to predictive fog data center migration. Fog computing, also called fog networking or fogging, describes a decentralized computing structure located between the cloud and devices that produce data. This flexible structure enables users to place resources, including applications and the data they produce, in logical locations to enhance performance. Fog computing bridges the gap between the cloud and Internet of Things (IoT) devices by enabling computing, storage, networking, and data management on the network nodes within the close vicinity of IoT devices. Therefore, computation, storage, networking, decision making, and data management occur along the path between IoT devices and the cloud, as data moves to the cloud from the IoT devices. Data center migration refers to migrating a data center to a new computing environment. Data center migration may include application migration, which refers to migrating one or more applications from one computing environment to another. Data center migration may also include data migration, which refers to migrating specific sets of data from one storage system to another. SUMMARY In a first aspect of the invention, there is a computer-implemented method including: obtaining, by a computing device, data items from data sources; classifying, by the computing device, the data items into categories using a first machine learning (ML) model; generating, by the computing device, a risk score of a first data center based on the classified data items and using a second machine learning (ML) model; determining, by the computing device, the risk score of the first data center exceeds a threshold; and in response to the determining the risk score of the first data center exceeds the threshold, initiating, by the computing device, a migration of the first data center to a second data center. In another aspect of the invention, there is a computer program product including one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media. The program instructions are executable to: obtain data items from data sources; classify the data items into categories using a first machine learning (ML) model; generate a risk score of a first data center based on the classified data items and using a second machine learning (ML) model; determine the risk score of the first data center exceeds a threshold; and in response to the determining the risk score of the first data center exceeds the threshold, initiate a migration of the first data center to a second data center. In another aspect of the invention, there is system including a processor, a computer readable memory, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media. The program instructions are executable to: obtain data items from data sources; generate respective risk scores of respective data centers based on the data items and using an ensemble machine learning framework; and in response to determining the risk score of a first one of the data centers exceeds a threshold, initiate migration of the first one of the data centers to a second one of the data centers based on the risk score of the second one of the data centers being less than the threshold. BRIEF DESCRIPTION OF THE DRAWINGS Aspects of the present invention are described in the detailed description which follows, in reference to the noted plurality of drawings by way of non-limiting examples of exemplary embodiments of the present invention. FIG. 1 depicts a cloud computing node according to an embodiment of the present invention. FIG. 2 depicts a cloud computing environment according to an embodiment of the present invention. FIG. 3 depicts abstraction model layers according to an embodiment of the present invention. FIG. 4 shows a block diagram of an exemplary environment in accordance with aspects of the invention. FIG. 5 shows a block diagram in accordance with aspects of the invention. FIG. 6 shows a flowchart of an exemplary method in accordance with aspects of the invention. DETAILED DESCRIPTION Aspects of the present invention relate generally to managing data centers and, more particularly, to predictive fog data center migration. Numerous events can negatively affect fog data center productivity or make a fog location unavailable and therefore affect the workloads at that fog site. One example of such events includes errors and failures within the fog data center, such as user errors in the data center, and power supply, network, storage, and compute failures in the data center. Another example involves disasters within the vicinity of a fog data center, such as natural disasters (e.g., earthquakes, volcanos, hurricanes, flooding, fire, drought, etc.) and non-natural disasters (e.g., bioh