Power management in disaggregated computing systems
For power management in a disaggregated computing system, a set of initial electrical power levels are allocated to a set of processor cores according to a predicted desired workload, where the set of initial power levels aggregate to an initial collective contracted power level. Electrical power is dynamically allocated to respective processor cores within the set of processor cores to produce a capacity to execute a collective demanded workload while maintaining the electrical power to the set of processor cores to an approximately constant electrical power level within a threshold of the initial collective contracted electrical power level.
1. A method for power management in a disaggregated computing system, by a processor device, comprising:
allocating a set of initial electrical power levels to a set of processor cores in the disaggregated computing system according to a predicted desired workload, wherein the set of initial power levels aggregate to an initial collective contracted power level; and
dynamically allocating electrical power to respective processor cores within the set of processor cores to produce a capacity to execute a collective demanded workload while maintaining the electrical power to the set of processor cores to an approximately constant electrical power level within a threshold of the initial collective contracted electrical power level;
and wherein the initial collective contracted power level comprises a total contracted electrical power level allocated from an electrical utility and the threshold comprises a contracted variation of an acceptable variable range above and below the total contracted electrical power level without incurring penalties from the electric utility.
2. The method of claim 1 , wherein the desired workload is comprised of respective workloads associated with the respective processor cores, the respective workloads associated with the respective processor cores being requested by a plurality of tenants.
3. The method of claim 2 , further including monitoring statistical information based on a workload throughput of the respective workloads associated with the respective processor cores to predict the desired workload.
4. The method of claim 2 , wherein a set of the plurality of tenants possess service level agreements (SLAs), and the electrical power is dynamically allocated to the respective processor cores according to the SLAs.
5. The method of claim 4 , further including dynamically allocating the electrical power according to a priority level of the SLAs, wherein respective SLAs have a higher priority than other SLAs.
6. The method of claim 2 , further including using a power management algorithm to compute, within a time window, a power model for each of the respective workloads associated with the respective processor cores.
7. The method of claim 6 , further including dynamically distributing the electrical power from one of the respective workloads to another one of the respective workloads using the power management algorithm, wherein the electrical power is dynamically distributed to maintain the approximately constant electrical power level within the threshold of the initial collective contracted electrical power level.
8. The method of claim 7 , further including iteratively updating the power model using the power management algorithm according to one of each respective workload and subsequent to an end of the time window.
9. A system for power management in a disaggregated computing system, the system comprising:
at least one processor device, wherein the at least one processor device:
allocates a set of initial electrical power levels to a set of processor cores in the disaggregated computing system according to a predicted desired workload, wherein the set of initial power levels aggregate to an initial collective contracted power level, and
dynamically allocates electrical power to respective processor cores within the set of processor cores to produce a capacity to execute a collective demanded workload while maintaining the electrical power to the set of processor cores to an approximately constant electrical power level within a threshold of the initial collective contracted electrical power level;
and wherein the initial collective contracted power level comprises a total contracted electrical power level allocated from an electrical utility and the threshold comprises a contracted variation of an acceptable variable range above and below the total contracted electrical power level without incurring penalties from the electric utility.
10. The system of claim 9 , wherein the desired workload is comprised of respective workloads associated with the respective processor cores, the respective workloads associated with the respective processor cores being requested by a plurality of tenants.
11. The system of claim 10 , wherein the at least one processor device monitors statistical information based on a workload throughput of the respective workloads associated with the respective processor cores to predict the desired workload.
12. The system of claim 10 , wherein a set of the plurality of tenants possess service level agreements (SLAs), and the electrical power is dynamically allocated to the respective processor cores according to the SLAs.
13. The system of claim 12 , wherein the at least one processor device dynamically allocates the electrical power according to a priority level of the SLAs, wherein respective SLAs have a higher priority than other SLAs.
14. The system of claim 10 , wherein the at least one processor device uses a power management algorithm to compute, within a time window, a power model for each of the respective workloads associated with the respective processor cores.
15. The system of claim 14 , wherein the at least one processor device dynamically distributes the electrical power from one of the respective workloads to another one of the respective workloads using the power management algorithm, wherein the electrical power is dynamically distributed to maintain the approximately constant electrical power level within the threshold of the initial collective contracted electrical power level.
16. The system of claim 15 , wherein the at least one processor device iteratively updates the power model using the power management algorithm according to one of each respective workload and subsequent to an end of the time window.
17. A computer program product for power management in a disaggregated computing system, by a processor device, 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 allocates a set of initial electrical power levels to a set of processor cores in the disaggregated computing system according to a predicted desired workload, wherein the set of initial power levels aggregate to an initial collective contracted power level; and
an executable portion that dynamically allocates electrical power to respective processor cores within the set of processor cores to produce a capacity to execute a collective demanded workload while maintaining the electrical power to the set of processor cores to an approximately constant electrical power level within a threshold of the initial collective contracted electrical power level;
and wherein the initial collective contracted power level comprises a total contracted electrical power level allocated from an electrical utility and the threshold comprises a contracted variation of an acceptable variable range above and below the total contracted electrical power level without incurring penalties from the electric utility.
18. The computer program product of claim 17 , wherein the desired workload is comprised of respective workloads associated with the respective processor cores, the respective workloads associated with the respective processor cores being requested by a plurality of tenants.
19. The computer program product of claim 18 , further including an executable portion that monitors statistical information based on a workload throughput of the respective workloads associated with the respective processor cores to predict the desired workload.
20. The computer program product of claim 18 , wherein a set of the plurality of tenants possess service level agreements (SLAs), and the electrical power is dynamically allocated to the respective processor cores according to the SLAs.
21. The computer program product of claim 20 , further including an executable portion that dynamically allocates the electrical power according to a priority level of the SLAs, wherein respective SLAs have a higher priority than other SLAs.
22. The computer program product of claim 18 , further including an executable portion that uses a power management algorithm to compute, within a time window, a power model for each of the respective workloads associated with the respective processor cores.
23. The computer program product of claim 22 , further including an executable portion that dynamically distributes the electrical power from one of the respective workloads to another one of the respective workloads using the power management algorithm, wherein the electrical power is dynamically distributed to maintain the approximately constant electrical power level within the threshold of the initial collective contracted electrical power level.
24. The computer program product of claim 23 , further including an executable portion that iteratively updates the power model using the power management algorithm according to one of each respective workload and subsequent to an end of the time window.