Service Processor Power Histogram for Out-of-Band Utilization Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power management systems in processing systems face challenges in estimating processor utilization without adding overhead to application processors and operating systems, especially under conditions of limited power availability and thermal management, and require operating system-dependent software and resources, which is not feasible during boot, shutdown, or maintenance.
Innovation Solution
A method and system that estimates processor utilization using power measurements made by a service processor, determining the idle power consumption and computing a threshold to calculate processor utilization as a fraction of maximum utilization, allowing for power management control without involving application or operating system information, by distributing power consumption over short intervals and using a histogram to separate idle and busy states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If operating system-dependent software is used to determine workloads and processing needs, then processor utilization estimation accuracy is improved, but system complexity and resource overhead increase
Solution Approach 1:
The patent extracts the processor utilization estimation function from the operating system environment and relocates it to the service processor. This is achieved by having the service processor directly monitor power consumption of application processors and map power levels to utilization percentages, eliminating the need for operating system-dependent software while maintaining estimation accuracy.
Solution Approach 2:
The patent introduces power consumption measurements as an intermediary parameter between actual processor utilization and the estimation mechanism. Instead of directly querying operating system resources, the service processor measures power consumption and uses predetermined mappings to estimate utilization, thereby avoiding operating system dependencies and reducing system complexity.
2Adaptability or versatility
If special software is installed to manage processor frequency and voltage, then power management capability is improved, but resource overhead and installation complexity increase
Solution Approach 1:
The service processor performs self-service by directly monitoring power consumption and autonomously determining processor utilization and appropriate power management actions. This eliminates the need for separate power management software installation, as the service processor uses its existing capabilities to measure power and map it to utilization levels, thereby improving ease of manufacture while maintaining adaptability.
3Productivity
If power management overhead is reduced to small portion of system processing, then system productivity is improved, but power management precision deteriorates
Solution Approach 1:
The patent replaces the mechanical/software-based approach of running power management tasks on application processors with an electrical measurement approach using the service processor's power monitoring capabilities. This substitution allows power management operations to occur without consuming application processor cycles, thereby maintaining system productivity while achieving fine-grained measurement precision through direct power consumption monitoring.
4Measurement precision
If power management tasks are performed on application processors, then measurement granularity is improved, but system productivity and energy consumption are negatively affected
Solution Approach 1:
The patent segments the system into two distinct functional domains: the service processor handles power management and utilization measurement tasks, while application processors focus solely on workload execution. This segmentation allows measurement granularity to be maintained through service processor monitoring without impacting application processor productivity or energy consumption, as the measurement functions are completely separated from the application processing domain.
Data Source
AI summary
A method and system for estimating processor utilization from power measurements provides an estimate of processor utilization that can be computed outside of the processor and operating system. Measurements of the processor power consumption are gathered over short intervals in a histogram. The idle power consumption of the processor is determined, and a threshold value higher than the idle power consumption level is computed from the idle power consumption. The number of histogram counts for bins greater than the threshold is normalized to the total number of measurements, providing a fractional value that corresponds to the processor utilization over the measurement interval. The fractional value can then be used in a power management algorithm that adjusts the frequency and optionally the voltage of the processor or group of processors based on their utilization.