Cloud-based compute resource configuration recommendation and allocation for database migration
Methods, systems, apparatuses, and computer-readable storage mediums described herein are directed to determining and recommending an optimal compute resource configuration for a cloud-based resource (e.g., a server, a virtual machine, etc.) for migrating a customer to the cloud. The embodiments described herein utilize a statistically robust approach that makes recommendations that are more flexible (elastic) and account for the full distribution of the amount of resource usage. Such an approach is utilized to develop a personalized rank of relevant recommendations to a customer. To determine which compute resource configuration to recommend to the customer, the customer's usage profile is matched to a set of customers that have already migrated to the cloud. The compute resource configuration that reaches the performance most similar to the performance of the configurations utilized by customers in the matched set is recommended to the user.
1. A system, comprising:
at least one processor circuit; and
at least one memory that stores program code configured to be executed by the at least one processor circuit, the program code comprising:
a configuration recommender configured to:
receive a plurality of time series of data values, each time series of the plurality of time series representative of a behavior of a respective metric, of a plurality of metrics, associated with a respective computing resource of an on-premise computing device during execution of an application on the on-premise computing device, the on-premise computing device associated with a first customer;
determine a group of second customers having a first level of similarity to the first customer;
generate a respective score for each cloud-based compute resource configuration of a plurality of cloud-based compute resource configurations of a cloud-based platform based on the plurality of time series, each score representing a probability that the application does not require throttling when executed on the respective cloud-based compute resource configuration, each of the cloud-based compute resource configurations utilized by a respective second customer of the group of second customers;
select a cloud-based compute resource configuration from the plurality of cloud-based compute resource configurations based at least on the scores; and
deploy the application on the selected cloud-based compute resource configuration.
2. The system of claim 1 , wherein the configuration recommender is further configured to:
for each cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations:
determine, for each metric of the plurality of metrics, a probability that the cloud-based compute resource configuration accommodates the behavior of the metric without throttling the application when executed on the cloud-based compute resource configuration; and
combine the probabilities to generate the score for the cloud-based compute resource configuration.
3. The system of claim 1 , wherein the configuration recommender is further configured to:
for each cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations, receive a respective cost value for utilizing the cloud-based compute resource configuration; and
determine candidate cloud-based compute resource configurations based on the scores generated for the plurality of cloud-based compute resource configurations and the cost values received for the plurality of cloud-based compute resource configurations, the candidate cloud-based compute resource configurations comprising a subset of cloud-based compute resource configurations from the plurality of cloud-based compute resource configurations.
4. The system of claim 3 , wherein the configuration recommender is further configured to:
determine the respective cloud-based compute resource configurations utilized by each of the second customers;
average the scores to generate an average score; and
select a cloud-based compute resource configuration from the candidate cloud-based compute resource configurations that is associated with a score having a second level of similarity to the average score.
5. The system of claim 4 , wherein the configuration recommender is further configured to:
provide a recommendation to the first customer recommending the selected cloud-based compute resource configuration;
receive user input that indicates that the first customer has designated the selected cloud-based compute resource configuration for migration; and
responsive to receiving the user input, deploy the application on the selected cloud-based compute resource configuration.
6. The system of claim 1 , wherein each cloud-based compute resource configuration of the plurality of cloud-based compute resource configuration comprises at least one of:
a number of central processing unit cores maintained by a respective cloud-based compute resource;
an amount of storage maintained by the respective cloud-based compute resource;
an amount to memory maintained by the respective cloud-based compute resource; or
a number of input/output operations per second supported by the respective cloud-based compute resource.
7. The system of claim 6 , wherein the respective cloud-based compute resource comprises at least one of:
a server maintained by the cloud-based platform; or
a virtual machine maintained by the cloud-based platform.
8. A computer-implemented method, comprising:
receiving a plurality of time series of data values, each time series of the plurality of time series representative of a behavior of a respective metric, of a plurality of metrics, associated with a respective computing resource of an on-premise computing device during execution of an application on the on-premise computing device;
for each cloud-based compute resource configuration of a plurality of cloud-based compute resource configurations of a cloud-based platform:
determining, for each metric of the plurality of metrics, a probability that the cloud-based compute resource configuration accommodates the behavior of the metric without throttling the application when executed on the cloud-based compute resource configuration, and
combining the probabilities to generate a score for the cloud-based compute resource configuration, the score representing a probability that the application does not require throttling when executed on the cloud-based compute resource configuration;
selecting a cloud-based compute resource configuration from the plurality of cloud-based compute resource configurations based at least on the scores; and
deploying the application on the selected cloud-based compute resource configuration.
9. The computer-implemented method of claim 8 , wherein said generating comprises:
for each cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations, receiving a respective cost value for utilizing the cloud-based compute resource configuration; and
determining candidate cloud-based compute resource configurations based on the scores generated for the plurality of cloud-based compute resource configurations and the cost values received for the plurality of cloud-based compute resource configurations, the candidate cloud-based compute resource configurations comprising a subset of cloud-based compute resource configurations from the plurality of cloud-based compute resource configurations.
10. The computer-implemented method of claim 9 , wherein said selecting comprises:
associating a customer of a cloud-based platform associated with the on-premise computing device with a group of customers of the cloud-based platform based on a first level of similarity therebetween;
determining a respective cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations utilized by each of the customers, the scores being generated for the determined cloud-based compute resource configurations;
averaging the scores to generate an average score; and
selecting a cloud-based compute resource configuration from the candidate cloud-based compute resource configurations that is associated with a score having a second level of similarity to the average score.
11. The computer-implemented method of claim 10 , wherein said deploying comprises:
providing a recommendation to the customer recommending the selected cloud-based compute resource configuration;
receiving user input that indicates that the customer has designated the selected cloud-based compute resource configuration for migration; and
responsive to receiving the user input, deploying the application on the selected cloud-based compute resource configuration.
12. The computer-implemented method of claim 8 , wherein each cloud-based compute resource configuration of the plurality of cloud-based compute resource configuration comprises at least one of:
a number of central processing unit cores maintained by a respective cloud-based compute resource;
an amount of storage maintained by the respective cloud-based compute resource;
an amount to memory maintained by the respective cloud-based compute resource; or
a number of input/output operations per second supported by the respective cloud-based compute resource.
13. The computer-implemented method of claim 12 , wherein the respective cloud-based compute resource comprises at least one of:
a server maintained by the cloud-based platform; or
a virtual machine maintained by the cloud-based platform.
14. A computer-readable storage medium having program instructions recorded thereon that, when executed by at least one processor, perform a method comprising:
receiving a plurality of time series of data values, each time series of the plurality of time series representative of a behavior of a respective metric, of a plurality of metrics, associated with a respective computing resource of an on-premise computing device during execution of an application on the on-premise computing device, the on-premise computing device associated with a first customer;
determine a group of second customers having a first level of similarity to the first customer;
generating a respective score for each cloud-based compute resource configuration of a plurality of cloud-based compute resource configurations of a cloud-based platform based on the plurality of time series, each score representing a probability that the application does not require throttling when executed on the respective cloud-based compute resource configuration, each of the cloud-based compute resource configurations utilized by a respective second customer of the group of second customers;
selecting a cloud-based compute resource configuration from the plurality of cloud-based compute resource configurations based at least on the scores; and
deploying the application on the selected cloud-based compute resource configuration.
15. The computer-readable storage medium of claim 14 , wherein said generating comprises:
for each cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations:
determining, for each metric of the plurality of metrics, a probability that the cloud-based compute resource configuration accommodates the behavior of the metric without throttling the application when executed on the cloud-based compute resource configuration; and
combining the probabilities to generate the score for the cloud-based compute resource configuration.
16. The computer-readable storage medium of claim 14 , wherein said generating comprises:
for each cloud-based compute resource configuration of the plurality of cloud-based compute resource configurations, receiving a respective cost value for utilizing the cloud-based compute resource configuration; and
determining candidate cloud-based compute resource configurations based on the scores generated for the plurality of cloud-based compute resource configurations and the cost values received for the plurality of cloud-based compute resource configurations, the candidate cloud-based compute resource configurations comprising a subset of cloud-based compute resource configurations from the plurality of cloud-based compute resource configurations.
17. The computer-readable storage medium of claim 16 , wherein the method further comprises:
determining the respective cloud-based compute resource configurations utilized by each of the second customers;
averaging the scores to generate an average score; and
selecting a cloud-based compute resource configuration from the candidate cloud-based compute resource configurations that is associated with a score having a second level of similarity to the average score.
18. The computer-readable storage medium of claim 17 , wherein said deploying comprises:
providing a recommendation to the first customer recommending the selected cloud-based compute resource configuration;
receiving user input that indicates that the first customer has designated the selected cloud-based compute resource configuration for migration; and
responsive to receiving the user input, deploying the application on the selected cloud-based compute resource configuration.
19. The computer-readable storage medium of claim 14 , wherein each cloud-based compute resource configuration of the plurality of cloud-based compute resource configuration comprises at least one of:
a number of central processing unit cores maintained by a respective cloud-based compute resource;
an amount of storage maintained by the respective cloud-based compute resource;
an amount to memory maintained by the respective cloud-based compute resource; or
a number of input/output operations per second supported by the respective cloud-based compute resource.
20. The system of claim 1 , wherein the configuration recommender is further configured to:
average the scores to generate an average score; and
select the cloud-based compute resource configuration from the plurality of cloud-based compute resource configurations based on a second level of similarity between the average score and the respective score of the selected cloud-based compute resource configuration.