Migrating workloads to a preferred environment
Migrating workloads to a preferred environment, including: predicting, for each of a plurality of environments, a performance load on each of a plurality of environments that would result from placing one or more of a plurality of workloads on the environment; determining a preferred environment for each of the plurality of workloads by determining a placement of each of the plurality of workloads that results in a best fit for the plurality of workloads; and deploying each of the plurality of workloads in the corresponding preferred environment.
1. A method comprising:
generating one or more predicted performance loads for one or more of a plurality of storage systems that would result from different permutations of workload migrations among the plurality of storage systems, using one or more machine learning models that are trained on workload characteristics for the plurality of storage systems;
based on the performance load predicted by the one or more machine learning models, determining one or more preferred workloads of a plurality of workloads to migrate and one or more preferred storage systems of the plurality of storage systems for receiving the one or more preferred workloads; and
migrating the one or more preferred workloads to the one or more preferred storage systems based on a permutation in which migrating the one or more preferred workloads to the one or more preferred storage systems provides a threshold predicted difference between performance capacity and predicted performance load in a storage system.
2. The method of claim 1 , wherein:
each workload includes a software application that is being executed and the software application accesses data that is stored by storage resources, wherein the software application is being executed in a first type of execution environment; and
deploying the workload in a preferred environment includes executing the software application in a second type of execution environment, wherein the first type of execution environment is different than the second type of execution environment.
3. The method of claim 2 , wherein the software application is a machine learning application.
4. The method of claim 2 , wherein the software application is an artificial intelligence application.
5. The method of claim 2 , wherein the first type of execution environment is a virtual machine and the second type of execution environment is a container.
6. The method of claim 2 , wherein deploying the workload in the preferred environment includes migrating the data that is accessed by the software application from a first storage environment to a second storage environment.
7. The method of claim 1 , wherein the one or more machine learning models are trained to identify correlations between an amount of performance load placed on a particular environment given characteristics of workloads executing on the particular environment at a same point in time.
8. The method of claim 1 , further comprising:
identifying, from amongst the plurality of workloads, a preferred workload to migrate; and
determining a preferred environment for each of the plurality of workloads by determining an optimal permutation of placement of the plurality of workloads on a plurality of environments that results in a largest reduction in performance load for at least one environment of the plurality of environments.
9. The method of claim 1 , wherein the one or more preferred workloads are executing prior to migration in a first environment that is an on-premises environment and wherein the one or more preferred environments are cloud environments.
10. The method of claim 1 , wherein the one or more preferred workloads are executing prior to migration in a cloud environment and the wherein the one or more preferred environments are on-premises environments.
11. A non-transitory computer readable storage medium storing instructions which, when executed, cause a processor to:
generate one or more predicted performance loads for one or more of a plurality of storage systems that would result from different permutations of workload migrations among the plurality of storage systems, using one or more machine learning models that are trained on workload characteristics for the plurality of storage systems;
based on the performance load predicted by the one or more machine learning models, determine one or more preferred workloads of a plurality of workloads to migrate and one or more preferred storage systems of the plurality of storage systems for receiving the one or more preferred workloads; and
migrate the one or more preferred workloads to the one or more preferred storage systems based on a permutation in which migrating the one or more preferred workloads to the one or more preferred storage systems provides a threshold predicted difference between performance capacity and predicted performance load in a storage system.
12. The non-transitory computer readable storage medium of claim 11 , wherein:
each workload includes a software application that is being executed and the software application accesses data that is stored by storage resources, wherein the software application is being executed in a first type of execution environment; and
deploying the workload in a preferred environment includes executing the software application in a second type of execution environment, wherein the first type of execution environment is different than the second type of execution environment.
13. The non-transitory computer readable storage medium of claim 12 , wherein deploying the workload in the preferred environment includes migrating the data that is accessed by the software application from a first storage environment to a second storage environment.
14. The non-transitory computer readable storage medium of claim 11 , wherein the one or more machine learning models are trained to identify correlations between an amount of performance load placed on a particular environment given characteristics of workloads executing on the particular environment at a same point in time.
15. The non-transitory computer readable storage medium of claim 11 , wherein the processor is further to:
identify, from amongst the plurality of workloads, a preferred workload to migrate.
16. An apparatus comprising:
a memory; and
a processor, operatively coupled to the memory, configured to:
generate one or more predicted performance loads for one or more of a plurality of storage systems that would result from different permutations of workload migrations among the plurality of storage systems, using one or more machine learning models that are trained on workload characteristics for the plurality of storage systems;
based on the performance load predicted by the one or more machine learning models, determine one or more preferred workloads of a plurality of workloads to migrate and one or more preferred storage systems of the plurality of storage systems for receiving the one or more preferred workloads; and
migrate the one or more preferred workloads to the one or more preferred storage systems based on a permutation in which migrating the one or more preferred workloads to the one or more preferred storage systems provides a threshold predicted difference between performance capacity and predicted performance load in a storage system.
17. The apparatus of claim 16 , wherein:
each workload includes a software application that is being executed and the software application accesses data that is stored by storage resources, wherein the software application is being executed in a first type of execution environment; and
deploying the workload in a preferred environment includes executing the software application in a second type of execution environment, wherein the first type of execution environment is different than the second type of execution environment.
18. The apparatus of claim 17 , wherein deploying the workload in the preferred environment includes migrating the data that is accessed by the software application from a first storage environment to a second storage environment.
19. The apparatus of claim 16 , wherein the one or more machine learning models are trained to identify correlations between an amount of performance load placed on a particular environment given characteristics of workloads executing on the particular environment at a same point in time.
20. The method of claim 2 , wherein the software application is a data analytics application.