Cloud-based monitoring of hardware components in a fleet of storage systems

1. A method comprising:

collecting, for a plurality of hardware components that are included in a physical storage system, information describing an operation of each hardware component, wherein information is collected for the hardware components of multiple physical storage systems;

predicting, based on the information describing the operation of each hardware component and historical information describing an operation of one or more other hardware components, an expected performance of each hardware component; and

modifying, based on the expected performance of each hardware component, a utilization of at least one or more of the physical storage systems in a fleet including utilizing at least one or more of the physical storage systems in the fleet in a limited service mode.

2. The method of claim 1 wherein predicting the expected performance of one or more other hardware components that are included in a physical storage system further comprises:

identifying a plurality of hardware components that were part of a same manufacturing batch; and

wherein the expected performance of hardware components that were part of the same manufacturing batch is predicted based on the information describing the operation of hardware components that were also part of the same manufacturing batch.

3. The method of claim 1 , wherein the hardware components are supercapacitors that are used to provide power to non-volatile memory in a write cache after a power failure.

4. The method of claim 3 , wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises decreasing the amount of data that can be stored in the write cache as the expected performance of the supercapacitors degrades.

5. The method of claim 1 , wherein the hardware components are QLC flash storage devices, and wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises storing data that is expected to remain valid for a period of time that is greater than a predetermined threshold on QLC flash storage devices whose expected longevity is below a predetermined longevity threshold.

6. The method of claim 1 , wherein the hardware components are power supplies and wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises distributing workloads across the fleet in dependence upon the expected performance of the power supplies.

7. The method of claim 1 , wherein the hardware components include one or more storage devices and the information describing the usage of the hardware components is consistent with an attack from a malicious actor.

8. The method of claim 1 , wherein modifying the utilization of at least one or more of the physical storage systems in the fleet includes utilizing at least one or more of the physical storage systems in the fleet in a reduced operating mode based on predicting a partial degradation of one or more hardware components.

9. An apparatus comprising a computer processor, a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out:

collecting, for a plurality of hardware components that are included in a physical storage system, information describing an operation of each hardware component, wherein information is collected for the hardware components of multiple physical storage systems;

predicting, based on the information describing the operation of each hardware component and historical information describing an operation of one or more other hardware components, an expected performance of each hardware component; and

modifying, based on the expected performance of each hardware component, a utilization of at least one or more of the physical storage systems in a fleet including utilizing at least one or more of the physical storage systems in the fleet in a limited service mode.

10. The apparatus of claim 9 , wherein predicting the expected performance of one or more other hardware components that are included in a physical storage system further comprises:

identifying a plurality of hardware components that were part of the same manufacturing batch; and

wherein the expected performance of hardware components that were part of a same manufacturing batch is predicted based on the information describing the operation of hardware components that were also part of the same manufacturing batch.

11. The apparatus of claim 9 , wherein the hardware components are supercapacitors that are used to provide power to non-volatile memory in a write cache after a power failure.

12. The apparatus of claim 11 , wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises decreasing the amount of data that can be stored in the write cache as the expected performance of the supercapacitors degrades.

13. The apparatus of claim 9 , wherein the hardware components are QLC flash storage devices, and wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises storing data that is expected to remain valid for a period of time that is greater than a predetermined threshold on QLC flash storage devices whose expected longevity is below a predetermined longevity threshold.

14. The apparatus of claim 9 , wherein the hardware components are power supplies and wherein modifying the utilization of at least one or more of the physical storage systems in the fleet further comprises distributing workloads across the fleet in dependence upon the expected performance of the power supplies.

15. The apparatus of claim 9 , wherein the hardware components include one or more storage devices and the information describing the usage of the hardware components is consistent with an attack from a malicious actor.

16. The apparatus of claim 9 , wherein modifying the utilization of at least one or more of the physical storage systems in the fleet includes utilizing at least one or more of the physical storage systems in the fleet in a reduced operating mode based on predicting a partial degradation of one or more hardware components.

17. A computer program product comprising a non-transitory computer readable storage medium with instructions embodied thereon that, when executed, are configured for:

receiving, for a plurality of hardware components that are included in a physical storage system, information describing an operation of each hardware component, wherein information is collected for the hardware components of multiple physical storage systems;

predicting, based on the information describing the operation each hardware component and historical information describing an operation of one or more other hardware components, an expected performance of each hardware component; and

modifying, based on the expected performance of each hardware component, a utilization of at least one or more of the physical storage systems in a fleet including utilizing at least one or more of the physical storage systems in the fleet in a limited service mode.

18. The computer program product of claim 17 , wherein the hardware components are supercapacitors that are used to provide power to non-volatile memory in a write cache after a power failure.

19. The computer program product of claim 18 , wherein initiating a modification to the utilization of at least one or more of the physical storage systems in the fleet further comprises initiating a decrease in the amount of data that can be stored in the write cache as the expected performance of the supercapacitors degrades.

20. The computer program product of claim 17 , wherein the computer program product is further configured for:

identifying a plurality of hardware components that were part of the same manufacturing batch; and

wherein the expected performance of hardware components that were part of a same manufacturing batch is predicted based on the information describing the operation of hardware components that were also part of the same manufacturing batch.