Resource lifecycle optimization in disaggregated data centers

1. A method for component lifecycle optimization in a disaggregated computing environment, by a processor, comprising:

performing a monitoring and machine learning process to learn a respective lifecycle of different resource types as the different resource types are assigned to respective workloads;

using the monitoring and machine learning process to develop a set of learned failure patterns for determining a mitigation action to perform as new faults are encountered within each of the different resource types while executing the respective workloads; and

performing the mitigation action to optimize a remaining lifecycle of respective ones of the different resource types according to the set of learned failure patterns, wherein the mitigation action includes:

detecting that a resource executing one of the respective workloads is exhibiting suspicious resource behavior;

responsive to detecting the suspicious resource behavior, transferring execution of the one of the respective executing workloads to an available resource from a resource pool of one of the different resource types;

instantiating a duplicate instance of the one of the respective executing workloads while emulating processing of the one of the respective executing workloads by the resource exhibiting the suspicious resource behavior; and

monitoring the resource exhibiting the suspicious resource behavior during the processing emulation by the monitoring and machine learning process to correlate the suspicious resource behavior to the learned failure patterns within a context of the one of the respective executing workloads.

2. The method of claim 1 , wherein the machine learning process is modified as the new faults are encountered within each of the different resource types.

3. The method of claim 1 , further including, pursuant to determining the mitigation action to perform, identifying one of the set of learned failure patterns based on a fault associated with one of the different resource types and a workload type which is executing on the one of the different resource types.

4. The method of claim 1 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on a service level agreement (SLA).

5. The method of claim 1 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on one of a projected lifetime of the workload (PLW) and an expected time to failure (ETTF) of the available resource.

6. The method of claim 1 , further including, commensurate with learning the respective lifecycle, storing data associated with the learned lifecycle for each of the different resource types within a resource tracking database; wherein the data includes health diagnostic information constituent to each respective resource of the different resource types.

7. The method of claim 6 , wherein the monitoring and machine learning process further includes monitoring each respective resource of the different resource types for the suspicious resource behavior, and correlating and aggregating results of the monitoring into one of a plurality of machine learning algorithms.

8. A system for component lifecycle optimization in a disaggregated computing environment, comprising:

a processor executing instructions stored in a memory device; wherein the processor:

performs a monitoring and machine learning process to learn a respective lifecycle of different resource types as the different resource types are assigned to respective workloads;

uses the monitoring and machine learning process to develop a set of learned failure patterns for determining a mitigation action to perform as new faults are encountered within each of the different resource types while executing the respective workloads; and

performs the mitigation action to optimize a remaining lifecycle of respective ones of the different resource types according to the set of learned failure patterns, wherein the mitigation action includes:

detecting that a resource executing one of the respective workloads is exhibiting suspicious resource behavior;

responsive to detecting the suspicious resource behavior, transferring execution of the one of the respective executing workloads to an available resource from a resource pool of one of the different resource types;

instantiating a duplicate instance of the one of the respective executing workloads while emulating processing of the one of the respective executing workloads by the resource exhibiting the suspicious resource behavior; and

monitoring the resource exhibiting the suspicious resource behavior during the processing emulation by the monitoring and machine learning process to correlate the suspicious resource behavior to the learned failure patterns within a context of the one of the respective executing workloads.

9. The system of claim 8 , wherein the machine learning process is modified as the new faults are encountered within each of the different resource types.

10. The system of claim 8 , wherein the processor, pursuant to determining the mitigation action to perform, identifies one of the set of learned failure patterns based on a fault associated with one of the different resource types and a workload type which is executing on the one of the different resource types.

11. The system of claim 8 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on a service level agreement (SLA).

12. The system of claim 8 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on one of a projected lifetime of the workload (PLW) and an expected time to failure (ETTF) of the available resource.

13. The system of claim 8 , wherein the processor, commensurate with learning the respective lifecycle, stores data associated with the learned lifecycle for each of the different resource types within a resource tracking database; wherein the data includes health diagnostic information constituent to each respective resource of the different resource types.

14. The system of claim 13 , wherein the monitoring and machine learning process further includes monitoring each respective resource of the different resource types for the suspicious resource behavior, and wherein the processor correlates and aggregates results of the monitoring into one of a plurality of machine learning algorithms.

15. A computer program product for component lifecycle optimization in a disaggregated computing environment, by a processor, the computer program product embodied on a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

an executable portion that performs a monitoring and machine learning process to learn a respective lifecycle of different resource types as the different resource types are assigned to respective workloads;

an executable portion that uses the monitoring and machine learning process to develop a set of learned failure patterns for determining a mitigation action to perform as new faults are encountered within each of the different resource types while executing the respective workloads; and

an executable portion that performs the mitigation action to optimize a remaining lifecycle of respective ones of the different resource types according to the set of learned failure patterns, wherein the mitigation action includes:

detecting that a resource executing one of the respective workloads is exhibiting suspicious resource behavior;

responsive to detecting the suspicious resource behavior, transferring execution of the one of the respective executing workloads to an available resource from a resource pool of one of the different resource types;

instantiating a duplicate instance of the one of the respective executing workloads while emulating processing of the one of the respective executing workloads by the resource exhibiting the suspicious resource behavior; and

monitoring the resource exhibiting the suspicious resource behavior during the processing emulation by the monitoring and machine learning process to correlate the suspicious resource behavior to the learned failure patterns within a context of the one of the respective executing workloads.

16. The computer program product of claim 15 , wherein the machine learning process is modified as the new faults are encountered within each of the different resource types.

17. The computer program product of claim 15 , further including an executable portion that, pursuant to determining the mitigation action to perform, identifies one of the set of learned failure patterns based on a fault associated with one of the different resource types and a workload type which is executing on the one of the different resource types.

18. The computer program product of claim 15 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on a service level agreement (SLA).

19. The computer program product of claim 15 , wherein the mitigation action further includes assigning the available resource from the resource pool of one of the different resource types to one of the respective executing workloads based on one of a projected lifetime of the workload (PLW) and an expected time to failure (ETTF) of the available resource.

20. The computer program product of claim 15 , further including an executable portion that, commensurate with learning the respective lifecycle, stores data associated with the learned lifecycle for each of the different resource types within a resource tracking database; wherein the data includes health diagnostic information constituent to each respective resource of the different resource types.

21. The computer program product of claim 20 , wherein the monitoring and machine learning process further includes an executable portion that monitors each respective resource of the different resource types for the suspicious resource behavior, and correlates and aggregates results of the monitoring into one of a plurality of machine learning algorithms.