Load Manager Threshold Logic for Computing Resource Power Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource pools face challenges in managing computational load and electrical power consumption while meeting service level agreements (SLAs), as the number of active resources affects both performance and cooling requirements, leading to inefficiencies and potential SLA violations.
Innovation Solution
A method and system that utilize multiple load thresholds to dynamically adjust the operational mode of computing resources, activating or deactivating resources based on load values and SLAs to optimize power consumption and maintain performance, including a Load Manager that compares load thresholds with actual values to change resource modes and ensure SLA compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of active computing resources is increased to meet computational load and SLA requirements, then service level agreement compliance is improved, but electrical power consumption and heat generation increase
Solution Approach 1:
The system dynamically adjusts the number of active computing resources based on varying computational load conditions. The load manager continuously monitors load values and compares them against thresholds to determine when to activate or deactivate resources, allowing the system to adapt its power consumption to actual workload requirements while maintaining SLA compliance.
Solution Approach 2:
The system changes the operational state parameter of computing resources from active to inactive based on load threshold comparisons. By monitoring load values and comparing them against predefined thresholds, the system transitions resources between operational states to optimize the balance between SLA compliance and power consumption.
2Productivity
If the number of active computing resources is increased to meet computational load, then productivity is improved, but cooling system power consumption increases
Solution Approach 1:
The system dynamically adjusts computing resources to match computational load requirements, which indirectly optimizes cooling system power consumption. By activating resources only when load values exceed thresholds and deactivating them when load decreases, the system reduces heat generation and corresponding cooling demands, thereby reducing cooling system energy consumption.
Solution Approach 2:
The system converts the harmful effect of heat generation into a beneficial control mechanism. By monitoring computational load and adjusting resource activation accordingly, the system uses load-induced heat generation as a signal to optimize cooling requirements, turning the thermal byproduct into a feedback mechanism for energy-efficient resource management.
3Loss of energy
If resources are frequently activated and deactivated to optimize power consumption, then electrical power efficiency is improved, but system stability deteriorates
Solution Approach 1:
The system establishes predefined load thresholds before operation begins. These thresholds serve as predetermined criteria that trigger resource activation or deactivation, preventing frequent or arbitrary state changes. By having thresholds set in advance, the system ensures that resources are only activated or deactivated when load conditions genuinely warrant such changes, thereby maintaining stability while optimizing power efficiency.
Solution Approach 2:
The load manager continuously monitors load values and compares them against thresholds to determine resource state changes. This feedback mechanism ensures that activation and deactivation decisions are based on actual system conditions rather than arbitrary timing, preventing unnecessary state transitions and maintaining system stability while achieving power optimization.
Data Source
AI summary
A method of managing a plurality of computing resources including obtaining a first load threshold and a second load threshold, obtaining a first load value and a second load value, comparing the first load threshold and the second load threshold with the first load value and the second load value, and changing an operating mode of a resource of the plurality of computing resources when both the first load threshold and the second load threshold are between the first load value and the second load value.