Dynanmic peak power limiting to processing nodes in an information handling system
A computer-implemented method dynamically limits peak power consumption in processing nodes of an IHS. A power management micro-controller receives processing node-level power-usage and workload data from several node controllers, including current power consumption and a current workload, for each processing node within the IHS. A total available system power of the IHS is identified including a peak power output capacity and a sustained output power capacity. At least one node peak power threshold is determined based on the power-usage and workload data for each of the processing nodes. The node controllers are triggered to determine and set a central processing unit (CPU) peak power limit for each of several CPUs within each of the processing nodes based on the node peak power threshold, wherein each of the CPUs dynamically adjusts an operating frequency based on the CPU peak power limit.
1. A computer-implemented method to dynamically limit peak power consumption in processing nodes of an information handling system (IHS), the method comprising:
receiving, at a power management micro-controller, a plurality of processing node-level power-usage and workload data from a plurality of node controllers, the data including a current power consumption and a current workload for each processing node within the IHS;
identifying a total available system power of the IHS, including a peak power output capacity and a sustained output power capacity;
determining at least one node peak power threshold based on the power-usage and workload data for each of the processing nodes; and
triggering the node controllers to determine and set a central processing unit (CPU) peak power limit for each of a plurality of CPUs within each of the processing nodes based on the node peak power threshold, wherein each of the CPUs dynamically adjusts an operating frequency based on the CPU peak power limit.
2. The method of claim 1 , further comprising:
determining at least one node average power threshold based on the power-usage and workload data for each of the processing nodes; and
transmitting the node peak power threshold and the node average power threshold to the node controllers of each of the processing nodes.
3. The method of claim 2 , further comprising:
triggering the node controllers to determine and set a CPU average power limit for each of the CPUs within each of the processing nodes based on the node average power threshold.
4. The method of claim 1 , further comprising:
initializing the power management micro-controller during start-up configuration of the IHS;
establishing communication between the power management micro-controller, the PSUs and the node controllers;
tracking, via the node controllers, the processing node power-usage and workload data for each of the processing nodes; and
triggering the node controllers to transmit the power-usage and workload data to the power management micro-controller.
5. The method of claim 1 , further comprising:
determining if the node peak power threshold has been exceeded for at least one of the processing nodes; and
in response to determining that the node peak power threshold has been exceeded for at least one of the processing nodes, instantaneously reducing the operating frequency of the CPUs within that processing node to a frequency that corresponds to a pre-determined power level.
6. The method of claim 5 , further comprising:
in response to determining that the node peak power threshold has not been exceeded for at least one of the processing nodes, determining if a current CPU peak power consumption is approaching the CPU peak power limit for at least one of the CPUs; and
in response to determining that the current CPU peak power consumption is approaching the CPU peak power limit, the corresponding CPUs estimating a current peak power level based on the workload data and adjusting the operating frequency of the CPUs such that the current peak power level remains less than the CPU peak power limit.
7. The method of claim 6 , further comprising:
in response to determining that the current CPU peak power consumption is not approaching the CPU peak power limit, determining if a current node average power consumption has exceeded the node average power threshold for at least one of the processing nodes; and
in response to determining that the current node average power consumption has exceeded the node average power threshold for at least one of the processing nodes, each corresponding CPU adjusting the operating frequency of the CPU such that the current average CPU power level is less than the CPU average power limit.
8. The method of claim 7 , further comprising:
in response to determining that the current node average power consumption has not exceeded the node average power threshold for at least one of the processing nodes, allowing the CPUs within that processing node to operate at a maximum allowable frequency.
9. The method of claim 1 , wherein the node controllers further include node power control firmware that execute on the node controllers to enable each of the node controllers to determine and set the CPU peak power limit.
10. The method of claim 1 , wherein the CPUs further include CPU power control firmware that execute on the CPUs to enable each of the CPUs to dynamically adjust the operating frequency of the CPU based on the CPU peak power limit.
11. The method of claim 1 , further comprising:
receiving at the power management micro-controller, a potential peak power range and a potential average power range from the processing nodes.
12. An information handling system (IHS) comprising:
one or more processing nodes, each of the processing nodes having at least one node controller, the node controllers each controlling node-level operations of the processing nodes within a corresponding node, the processing nodes each having at least one central processing unit (CPU) communicatively coupled to the node controller;
a power management module having a power management micro-controller and a memory coupled to the power management micro-controller, the power management module communicatively coupled to the processing nodes;
a power subsystem communicatively coupled to the power management micro-controller and providing power distribution to the IHS, the power subsystem including at least one power supply unit (PSU); and
the power management micro-controller having firmware executing thereon to enable dynamic peak power limiting to the processing nodes within the IHS, wherein the firmware configures the power management micro-controller to:
receive a plurality of processing node-level power-usage and workload data from the node controllers, the data including a current power consumption and a current workload for each processing node within the IHS;
identify a total available system power of the IHS including a peak power output capacity and a sustained output power capacity;
determine at least one node peak power threshold based on the power-usage and workload data for each of the processing nodes; and
trigger the node controllers to determine and set a CPU peak power limit for each of the CPUs within each of the processing nodes based on the node peak power threshold, wherein each of the CPUs dynamically adjusts an operating frequency based on the CPU peak power limit.
13. The information handling system of claim 12 , wherein the firmware further configures the power management micro-controller to:
determine at least one average node power threshold based on the power-usage and workload data for each of the processing nodes; and
transmit the node peak power threshold and the node average power threshold to the node controllers at each of the processing nodes.
14. The information handling system of claim 13 wherein the firmware further configures the power management micro-controller to:
trigger the node controllers to determine and set a CPU average power limit for each of the CPUs within each of the processing nodes based on the node average power threshold.
15. The information handling system of claim 12 wherein the firmware further configures the power management micro-controller to:
initialize the power management micro-controller during start-up configuration of the IHS;
establish communication between the power management micro-controller, the PSUs and the node controllers;
track, via the node controllers, the processing node power-usage and workload data for each of the processing nodes; and
trigger the node controllers to transmit the power-usage and workload data to the power management micro-controller.
16. The information handling system of claim 12 wherein the node controllers have node controller power control firmware executing thereon and the node controller firmware configures the node controllers to:
determine if the node peak power threshold has been exceeded for at least one of the processing nodes; and
in response to determining that the node peak power threshold has been exceeded for at least one of the processing nodes, instantaneously reducing the operating frequency of the CPUs within that processing node to a frequency that corresponds to a pre-determined power level.
17. The information handling system of claim 16 wherein the CPUs have CPU power control firmware executing thereon and the firmware configures the CPUs to:
in response to determining that the node peak power threshold has not been exceeded for at least one of the processing nodes, determine if a current CPU peak power consumption is approaching the CPU peak power limit for at least one of the CPUs; and
in response to determining that the current CPU peak power consumption is approaching the CPU peak power limit, the corresponding CPUs estimate a current peak power level based on the workload data and adjust the operating frequency of the CPUs such that the current peak power level remains less than the CPU peak power limit.
18. The information handling system of claim 17 wherein the CPUs have CPU power control firmware executing thereon and the firmware configures the CPUs to:
in response to determining that the current CPU peak power consumption is not approaching the CPU peak power limit, determine if a current node average power consumption has exceeded the node average power threshold for at least one of the processing nodes; and
in response to determining that the current node average power consumption has exceeded the node average power threshold for at least one of the processing nodes, each corresponding CPU adjusts the operating frequency of the CPU such that the current average CPU power level is less than the CPU average power limit.
19. The information handling system of claim 18 wherein the CPUs have CPU power control firmware executing thereon and the firmware configures the CPUs to:
in response to determining that the current node average power consumption has not exceeded the node average power threshold for at least one of the processing nodes, operate the CPUs within that processing node at a maximum allowable frequency.
20. The information handling system of claim 12 wherein the firmware further configures the power management micro-controller to:
receive a potential peak power range and a potential average power range from the processing nodes.