Optimized auto-tiering, wherein subset of data movements are selected, utilizing workload skew point, from a list that ranks data movements based on criteria other than I/O workload

1. A method of performing data movement optimization processing comprising:

receiving, using a processor, a list of proposed data movements for a plurality of data portions, wherein each proposed data movement of the list denotes a proposed movement of a data portion of the plurality of data portions from a source storage tier of a plurality of storage tiers to a target tier of the plurality of storage tiers, wherein the plurality of storage tiers are used for back-end data storage in a data storage system, and wherein each proposed data movement of the list denotes either a demotion, where the source storage tier is a higher performance tier than the target storage tier, or a promotion, where the source storage tier is a lower performance tier than the target storage tier;

ranking, using a processor, the list of proposed data movements in accordance with one or more criteria associated with each of the proposed data movements of the list, wherein the one or more criteria for said each proposed data movement includes at least one characteristic about said each proposed data movement other than I/O workload directed to the data portion of said each proposed data movement, wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an indicator denoting whether the data portion includes file system metadata, wherein the file system metadata includes protection or access information indicating user permissions needed to access different data portions;

selecting, using a processor and in accordance with said ranking, a subset of the proposed data movements of the list, wherein the subset of the proposed data movements of the list is determined using a workload skew point denoting that a specified percentage of I/O workload of the plurality of data portions is directed to a specified percentage of busiest data portions of the plurality of data portions;

implementing, using a processor, the subset of the proposed data movements by performing each of the proposed data movements of the subset;

and revising, using a processor, the list to remove the subset of the proposed data movements implemented in said implementing.

2. The method of claim 1 , wherein the plurality of storage tiers are ranked based on performance.

3. The method of claim 2 , wherein the list of proposed data movements is generated by a data storage optimizer that automatically generates a new list of proposed data movements at each occurrence of a defined time period.

4. The method of claim 3 , wherein said ranking, said selecting, said implementing and said revising are repeated as a group at each occurrence of a defined time period within a cycle, wherein the cycle has a defined cycle time that is a multiple of the defined time period.

5. The method of claim 1 , wherein the one or more criteria include one or more primary criteria and one or more secondary criteria, and wherein each proposed data movement of the list meeting the one or more primary criteria is ranked higher by said ranking than any other proposed data movement of the list not meeting the primary criteria.

6. The method of claim 1 , wherein the one or more criteria includes a third level of criteria.

7. The method of claim 1 , wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an indicator denoting whether the data portion includes application critical data.

8. The method of claim 1 , wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an I/O workload for the data portion.

9. The method of claim 8 , wherein the I/O workload for the data portion denotes a cumulative I/O workload directed to the data portion for both reads and writes directed to the data portion.

10. The method of claim 1 , wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes one of a predefined set of application priorities denoting an application priority for an application using the data included in the data portion of said each proposed data movement.

11. The method of claim 1 , wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes one of a predefined set of quality of service levels denoting a priority or a level of importance of the data portion of said each proposed data movement, wherein the priority or the level of importance of the data portion of said each proposed data movement is with respect to an application using the data included in the data portion of said each proposed data movement.

12. The method of claim 1 , wherein a first of the plurality of storage tiers includes physical storage devices having a first write endurance and a second of the plurality of storage tiers includes physical storage devices having a second write endurance greater than the first write endurance.

13. The method of claim 12 , wherein the first endurance denotes that physical storage devices of the first storage tier wear out and are expected to sustain an expected number of write or erasure cycles that is less than a second expected number of write or erasure cycles associated with physical devices of the second storage tier.

14. The method of claim 13 , wherein the first and the second storage tiers are tiers of different flash-based physical drives.

15. The method of claim 14 , wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an indicator denoting whether I/O workload directed to the data portion is read heavy or write heavy.

16. The method of claim 1 , wherein the subset of the proposed data movements is further determined using a specified increase in performance expected as a result of modeling implementation of a number of the proposed data movements of the list as ranked by said ranking.

17. A system comprising:

one or more processors;

and one or more memories comprising code stored thereon that, when executed, performs a method comprising:

receiving, using a processor, a list of proposed data movements for a plurality of data portions, wherein each proposed data movement of the list denotes a proposed movement of a data portion of the plurality of data portions from a source storage tier of a plurality of storage tiers to a target tier of the plurality of storage tiers, wherein the plurality of storage tiers are used for back-end data storage in a data storage system, and wherein each proposed data movement of the list denotes either a demotion, where the source storage tier is a higher performance tier than the target storage tier, or a promotion, where the source storage tier is a lower performance tier than the target storage tier;

ranking, using a processor, the list of proposed data movements in accordance with one or more criteria associated with each of the proposed data movements of the list, wherein the one or more criteria for said each proposed data movement includes at least one characteristic about said each proposed data movement other than I/O workload directed to the data portion of said each proposed data movement, wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an indicator denoting whether the data portion includes file system metadata, wherein the file system metadata includes protection or access information indicating user permissions needed to access different data portions;

selecting, using a processor and in accordance with said ranking, a subset of the proposed data movements of the list, wherein the subset of the proposed data movements of the list is determined using a workload skew point denoting that a specified percentage of I/O workload of the plurality of data portions is directed to a specified percentage of busiest data portions of the plurality of data portions;

implementing, using a processor, the subset of the proposed data movements by performing each of the proposed data movements of the subset;

and revising, using a processor, the list to remove the subset of the proposed data movements implemented in said implementing.

18. A non-transitory computer readable medium comprising code stored thereon, that when executed, performs a method of data movement optimization processing comprising:

receiving, using a processor, a list of proposed data movements for a plurality of data portions, wherein each proposed data movement of the list denotes a proposed movement of a data portion of the plurality of data portions from a source storage tier of a plurality of storage tiers to a target tier of the plurality of storage tiers, wherein the plurality of storage tiers are used for back-end data storage in a data storage system, and wherein each proposed data movement of the list denotes either a demotion, where the source storage tier is a higher performance tier than the target storage tier, or a promotion, where the source storage tier is a lower performance tier than the target storage tier;

ranking, using a processor, the list of proposed data movements in accordance with one or more criteria associated with each of the proposed data movements of the list, wherein the one or more criteria for said each proposed data movement includes at least one characteristic about said each proposed data movement other than I/O workload directed to the data portion of said each proposed data movement, wherein the one or more criteria for said each proposed data movement, that moves the data portion to the target storage tier, includes an indicator denoting whether the data portion includes file system metadata, wherein the file system metadata includes protection or access information indicating user permissions needed to access different data portions;

selecting, using a processor and in accordance with said ranking, a subset of the proposed data movements of the list, wherein the subset of the proposed data movements of the list is determined using a workload skew point denoting that a specified percentage of I/O workload of the plurality of data portions is directed to a specified percentage of busiest data portions of the plurality of data portions;

implementing, using a processor, the subset of the proposed data movements by performing each of the proposed data movements of the subset;

and revising, using a processor, the list to remove the subset of the proposed data movements implemented in said implementing.

19. The non-transitory computer readable medium of claim 18 , wherein the subset includes a highest ranked portion of proposed data movements on the list, wherein the revising step outputs a revised list of proposed data movements in which the subset implemented have been removed, and, after performing said revising, the method further includes performing:

receiving an on-demand request to perform data movements; and

responsive to receiving the on-demand request, implementing the revised list of proposed data movements.

20. The non-transitory computer readable medium of claim 18 , wherein the subset of the proposed data movements is further determined using a specified increase in performance expected as a result of modeling implementation of a number of the proposed data movements of the list as ranked by said ranking.