Running average power level assignment in an information handling system

Patent metadata

Field	Value
Publication number	US-2025138610-A1
Application number	US-202318500022-A
Country	US
Kind code	A1
Filing date	Nov 1, 2023
Priority date	Nov 1, 2023
Publication date	May 1, 2025
Grant date	—

Abstract

An information handling system stores a knob table, and determines time series data for one or more applications being executed. The information handling system extracts features associated with the system, and determines a system type for the information handling system. The system determines a running average power level based on the time series data, the extracted features, and the system type. Based on the determined running average power level, the information handling system determines a configuration knob to optimize the system. The configuration knob is determined from the knob table in the memory.

First claim

What is claimed is:

1 . An information handling system comprising:
a memory to store a knob table; and a processor to communicate with the memory, wherein during a runtime of the information handling system, the processor to:
determine time series data for one or more applications being executed;
extract features associated with the information handling system;
determine a system type for the information handling system;
determine a running average power level based on the time series data, the extracted features, and the system type; and
determine a configuration knob to optimize the information handling system based on the determined running average power level, wherein the configuration knob is determined from the knob table in the memory.

2 . The information handling system of claim 1 , wherein the determination of the configuration knob includes the processor further to:
determine an application to be optimized in the information handling system; and determine a user selected knob priority, wherein the determination of the configuration knob is further determined based on the application to the optimized and the user selected knob priority.

3 . The information handling system of claim 2 , wherein the determination of the configuration knob is further based on a utilization of the application to be optimized.

4 . The information handling system of claim 3 , wherein the utilization of the application includes data associated with system resource allocations exclusive to the application.

5 . The information handling system of claim 1 , wherein prior to the runtime, the processor further to: execute a clustering model to group information handling systems into several different system types based on features associated with the information handling systems.

6 . The information handling system of claim 5 , wherein for each system type of the several different system types and prior to the runtime, the processor further to:
execute second clustering model to generate a plurality of running average power level clusters; and assign a high level, a medium level, or a low level to each of the running average power levels.

7 . The information handling system of claim 1 , wherein the determination of the running average power level includes the processor further to: assign the information handling system to a running average power level cluster.

8 . The information handling system of claim 1 , wherein time series data includes physical parameters of the information handling system, types of the applications, utilization of the applications, and application concurrency data.

9 . A method comprising:
during a runtime of an information handling system:
determining, by a processor of the information handling system, time series data for one or more applications being executed;
extracting features associated with the information handling system;
determining a system type for the information handling system;
determining a running average power level based on the time series data, the extracted features, and the system type; and
determining, by the processor, a configuration knob to optimize the information handling system based on the determined running average power level, wherein the configuration knob is determined from the knob table in the memory.

10 . The method of claim 9 , wherein the determining of the configuration knob includes the method further comprising:
determining an application to be optimized in the information handling system; and determining a user selected knob priority, wherein the determination of the configuration knob is further determined based on the application to the optimized and the user selected knob priority.

11 . The method of claim 10 , wherein the determining of the configuration knob is further based on a utilization of the application to be optimized.

12 . The method of claim 11 , wherein the utilization of the application includes data associated with system resource allocations exclusive to the application.

13 . The method of claim 12 , wherein prior to the runtime, the method further comprising: execute a clustering model to group information handling systems into several different system types based on features associated with the information handling systems.

14 . The method of claim 13 , wherein for each system type of the several different system types and prior to the runtime, the method further comprising:
executing second clustering model to generate a plurality of running average power level clusters; and assigning a high level, a medium level, or a low level to each of the running average power levels.

15 . The method of claim 9 , wherein the determination of the running average power level includes the processor further to: assigning the information handling system to a running average power level cluster.

16 . The method of claim 9 , wherein time series data includes physical parameters of the information handling system, types of the applications, utilization of the applications, and application concurrency data.

17 . An information handling system comprising:
a memory to store a knob table; and a processor to:
execute a first clustering model to generate a plurality of running average power level clusters;
assign a high level, a medium level, or a low level to each of the running average power levels;
during a runtime of the information handling system, the processor to:
determine time series data for one or more applications being executed;
extract features associated with the information handling system;
determine a system type for the information handling system;
determine a running average power level based on the time series data, the extracted features, and the system type;
assign the information handling system to a running average power level cluster; and
determine a configuration knob to optimize the information handling system based on the determined running average power level, wherein the configuration knob is determined from the knob table in the memory.

18 . The information handling system of claim 17 , wherein the determination of the configuration knob includes the processor further to:
determine an application to be optimized in the information handling system; and determine a user selected knob priority, wherein the determination of the configuration knob is further determined based on the application to the optimized and the user selected knob priority.

19 . The information handling system of claim 18 , wherein the determination of the configuration knob is further based on a utilization of the application to be optimized.

20 . The information handling system of claim 19 , wherein the utilization of the application includes data associated with system resource allocations exclusive to the application.

Assignees

• Dell Products Lp

Inventors

• Khosrowpour Farzad
• Ning Yan

Classifications

• G06F1/26 Primary

  Power supply means, e.g. regulation thereof (for memories G11C) · CPC title