Dynamic power capping of multi-server nodes in a chassis based on real-time resource utilization
US-2020012334-A1 · Jan 9, 2020 · US
US11397457B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-11397457-B2 |
| Application number | US-202016914080-A |
| Country | US |
| Kind code | B2 |
| Filing date | Jun 26, 2020 |
| Priority date | Jun 26, 2020 |
| Publication date | Jul 26, 2022 |
| Grant date | Jul 26, 2022 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
One aspect of the present disclosure involves dynamically performing power capping with respect to a group of computing systems. Different priority levels can be assigned to at least some of the individual computing systems within the group of computing systems. Individual power limits can be set for the plurality of individual computing systems based at least in part on the different priority levels and utilization levels of the plurality of individual computing systems. Another aspect of the present disclosure involves dynamically performing power capping with respect to various subsystems of a computing system. Different priority levels can be assigned to at least some of the plurality of individual subsystems within the computing system. Individual power limits can be set for the plurality of individual subsystems based at least in part on the different priority levels and current power consumption of the plurality of individual subsystems.
Opening claim text (preview).
What is claimed is: 1. A method for dynamically performing power capping with respect to a group of computing systems, the group of computing systems comprising a plurality of individual computing systems, the method comprising: assigning different priority levels to the plurality of individual computing systems within the group of computing systems based at least in part on types of processing resources associated with running different types of workloads on individual computing systems form the plurality of individual computing systems; determining a total power limit for the group of computing systems; determining utilization levels of the plurality of individual computing systems within the group of computing systems; setting individual power limits for the plurality of individual computing systems within the group of computing systems based at least in part on the total power limit, the different priority levels, and the utilization levels, wherein an individual power limit for an individual computing system indicates a total amount of power that the individual computing system is permitted to use; and causing power consumption of the plurality of individual computing systems within the group of computing systems to be adjusted to satisfy the individual power limits. 2. The method of claim 1 , wherein: assigning the different priority levels comprises assigning a first priority level to a individual first computing system and assigning a second priority level to a second individual computing system; setting the individual power limits comprises setting a first power limit for the first individual computing system and setting a second power limit for the second individual computing system; the first priority level is higher than the second priority level; and the first power limit is higher than the second power limit. 3. The method of claim 1 , wherein: assigning the different priority levels comprises assigning a first priority level to a first individual computing system and assigning a second priority level to a individual second computing system; determining the utilization levels comprises determining a first utilization level of the first individual computing system and determining a second utilization level of the second individual computing system; setting the individual power limits comprises setting a first power limit for the first individual computing system and setting a second power limit for the second individual computing system; the first priority level is equal to the second priority level; the first utilization level is higher than the second utilization level; and the first power limit is higher than the second power limit. 4. The method of claim 1 , wherein assigning the different priority levels comprises receiving user input specifying at least some of the different priority levels for the plurality of individual computing systems, or wherein assigning the different priority levels comprises automatically determining at least some of the different priority levels based at least in part on historical information about the plurality of individual computing systems within the group of computing systems. 5. The method of claim 1 , wherein assigning the different priority levels includes: assigning a first priority level to a first computing system from the plurality of individual computing systems based on the first computing system having a processor-intensive workload; and assigning a second priority level to a second computing system from the plurality of individual computing systems based on the second computing system having a storage-intensive workload. 6. The method of claim 5 , wherein setting the individual power limits for the plurality of individual computing systems includes: assigning a higher CPU limit for the first computing system than the second computing system based on the first computing system having the processor-intensive workload; and assigning a higher storage limit for the second computing system than the first computing system based on the second computing system having the storage-intensive workload. 7. A method for dynamically performing power capping with respect to a plurality of individual subsystems within a computing system, comprising: assigning different priority levels to the plurality of individual subsystems within the computing system based at least in part on types of processing resources associated with running different types of workloads on individual subsystems form the plurality of individual subsystems; determining a total power limit for the computing system; determining current power consumption of the plurality of individual subsystems within the computing system; setting individual power limits for the plurality of individual subsystems within the computing system based at least in part on the different priority levels, the total power limit, and the current power consumption, wherein an individual power limit for an individual subsystem indicates a total amount of power that the individual subsystem is permitted to use; and causing power consumption of the plurality of individual subsystems within the computing system to be adjusted to satisfy the individual power limits. 8. The method of claim 7 , wherein: assigning the different priority levels comprises assigning a first priority level to a first subsystem and assigning a second priority level to a second subsystem; setting the individual power limits comprises setting a first power limit for the first subsystem and setting a second power limit for the second subsystem; the first priority level is higher than the second priority level; and the first power limit is higher than the second power limit. 9. The method of claim 7 , wherein: assigning the different priority levels comprises assigning a first priority level to a first subsystem and assigning a second priority level to a second subsystem; determining the current power consumption comprises determining a first power consumption of the first subsystem and determining a second power consumption of the second subsystem; setting the individual power limits comprises setting a first power limit for the first subsystem and setting a second power limit for the second subsystem; the first priority level is equal to the second priority level; the first power consumption is higher than the second power consumption; and the first power limit is higher than the second power limit. 10. The method of claim 7 , wherein assigning the different priority levels comprises receiving user input specifying at least some of the different priority levels for the plurality of individual subsystems, or wherein assigning the different priority levels comprises automatically determining at least some of the different priority levels based at least in part on historical information about the plurality of individual subsystems. 11. The method of claim 7 , further comprising: determining a plurality of weights corresponding to the plurality of individual subsystems within the computing system, wherein the plurality of weights depend at least in part on the different priority levels that have been assigned to the plurality of individual subsystems; and applying the plurality of weights to a proportional difference between the total power limit for the computing system and a total current power consumption of the computing system. 12. The method of claim 7 , wherein assigning the different priority levels includes: assigning a first priority level to a first subsystem from the plurality of individual subsystems within the computing system based on the first subsystem
Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title
Blade assemblies, e.g. blade cases or inner arrangements within a blade · CPC title
Power management, i.e. event-based initiation of a power-saving mode · CPC title
Monitoring of events, devices or parameters that trigger a change in power modality · CPC title
Monitoring task completion, e.g. by use of idle timers, stop commands or wait commands · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.