Selecting a storage array for data storage based on anticipated energy savings for that storage array

1 . A method comprising:

determining an anticipated energy cost for transmitting certain communications between a trustworthy energy awareness (TEA) module, a storage controller, and at least one storage array included among a plurality of storage arrays, wherein the certain communications are ones that are structured to identify a most energy-efficient storage array, as compared to other storage arrays included in the plurality of storage arrays;

determining an expected energy savings that is likely to occur as a result of using the most energy-efficient storage array to store a set of data;

determining whether the anticipated energy cost exceeds the expected energy savings;

in response to determining that the anticipated energy cost does exceed the energy savings, forgoing transmitting the certain communications so as to avoid expending the anticipated energy cost and instead storing the set of data in a selected storage array; and

in response to determining that the anticipated energy cost does not exceed the energy savings:

receiving, by the TEA module, a request from the storage controller to identify the most energy-efficient storage array on which to store the set of data;

determining, by the TEA module based on energy efficiency information and user constraints, the most energy-efficient storage array;

identifying, by the TEA module to the storage controller, the most energy-efficient storage array; and

storing the set of data on the most energy-efficient storage array.

2 . The method as recited in claim 1 , wherein the determining of the most energy-efficient storage array is performed using a machine learning (ML) model of the TEA module.

3 . The method as recited in claim 1 , wherein the most energy-efficient storage array is determined after the user constraints have been met.

4 . The method as recited in claim 1 , wherein the energy efficiency information comprises historical energy efficiency information and/or predicted respective energy efficiencies of the storage arrays.

5 . The method as recited in claim 1 , wherein the energy efficiency information is obtained from the storage arrays.

6 . The method as recited in claim 1 , wherein the data is a chunk of data.

7 . The method as recited in claim 1 , wherein the receiving, the determining, and the identifying, are each performed for each chunk of data in a group of chunks of data.

8 . The method as recited in claim 1 , wherein the storage arrays are elements of a distributed storage environment that spans multiple data owners and storage providers.

9 . The method as recited in claim 2 , wherein the ML model is a trained ML model that was trained using historical information including a respective energy efficiency of one or more of the storage arrays.

10 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising:

determining an anticipated energy cost for transmitting certain communications between a trustworthy energy awareness (TEA) module, a storage controller, and at least one storage array included among a plurality of storage arrays, wherein the certain communications are ones that are structured to identify a most energy-efficient storage array, as compared to other storage arrays included in the plurality of storage arrays;

determining an expected energy savings that is likely to occur as a result of using the most energy-efficient storage array to store a set of data;

determining whether the anticipated energy cost exceeds the expected energy savings;

in response to determining that the anticipated energy cost does exceed the energy savings, forgoing transmitting the certain communications so as to avoid expending the anticipated energy cost and instead storing the set of data in a selected storage array; and

in response to determining that the anticipated energy cost does not exceed the energy savings:

receiving, by the TEA module, a request from the storage controller to identify the most energy-efficient storage array on which to store the set of data;

determining, by the TEA module based on energy efficiency information and user constraints, the most energy efficient storage array;

identifying, by the TEA module to the storage controller, the most energy-efficient storage array; and

storing the set of data on the most energy-efficient storage array.

11 . The non-transitory storage medium as recited in claim 10 , wherein the determining of the most energy-efficient storage array is performed using a machine learning (ML) model of the TEA module.

12 . The non-transitory storage medium as recited in claim 10 , wherein the most energy-efficient storage array is determined after the user constraints have been met.

13 . The non-transitory storage medium as recited in claim 10 , wherein the energy efficiency information comprises historical energy efficiency information and/or predicted respective energy efficiencies of the storage arrays.

14 . The non-transitory storage medium as recited in claim 10 , wherein the energy efficiency information is obtained from the storage arrays.

15 . The non-transitory storage medium as recited in claim 10 , wherein the data is a chunk of data.

16 . The non-transitory storage medium as recited in claim 10 , wherein the receiving, the determining, and the identifying, are each performed for each chunk of data in a group of chunks of data.

17 . The non-transitory storage medium as recited in claim 10 , wherein the storage arrays are elements of a distributed storage environment that spans multiple data owners and storage providers.

18 . The non-transitory storage medium as recited in claim 11 , wherein the ML model is a trained ML model that was trained using historical information including a respective energy efficiency of one or more of the storage arrays.