Data Center Power Efficiency Calculation With Weighted Task Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating Data Center energy Productivity (DCeP) do not consider the weight for each application type or importance of processing between tasks, leading to inappropriate power efficiency calculations.
Innovation Solution
A power efficiency calculation device that collects metrics, normalizes task execution amounts, and sets importance groups with weight coefficients to reflect the importance of applications and tasks, using a metrics collection unit, task execution amount normalization unit, and importance group setting unit to calculate DCeP.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If all tasks are handled equivalently without weight coefficients, then the calculation is simple, but the power efficiency cannot reflect the importance of different applications and tasks
Solution Approach 1:
The patent applies local quality by assigning different weight coefficients to different tasks based on their importance. Instead of treating all tasks equally, the system assigns higher weights to more important tasks (e.g., critical business operations) and lower weights to less important tasks, thereby reflecting the local quality differences in task importance within the overall power efficiency calculation.
Solution Approach 2:
The patent changes the parameter of task importance by introducing weight coefficients as a new parameter. The power efficiency calculation transitions from a simple count of task executions to a weighted sum that incorporates importance levels, thereby accurately reflecting the impact of different tasks on overall system performance and power efficiency.
2Measurement precision
If weight coefficients are introduced for different application types and tasks, then the power efficiency calculation reflects importance accurately, but the calculation complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-defining weight coefficients for different task types and application categories before the actual power efficiency calculation. The system establishes the importance levels and corresponding weights in advance, so that during the calculation phase, the system simply needs to multiply task execution counts by their pre-determined weights, thereby reducing real-time calculation complexity while maintaining measurement accuracy.
3Reliability
If the assessment window is extended to capture more task executions, then the measurement becomes more reliable, but the real-time calculation capability is reduced
Solution Approach 1:
The patent applies feedback by continuously monitoring task executions and updating the power efficiency calculation in real-time. The system collects metrics from multiple assessment windows and uses feedback loops to maintain an up-to-date view of task execution patterns, allowing the system to provide reliable measurements without requiring excessively long assessment windows, thereby balancing reliability with calculation speed.
Data Source
AI summary
A power efficiency calculation device includes a metrics collection unit that collects metrics from a physical server group, a task execution amount normalization unit that determines a normalization coefficient for each task of each application, an importance group setting unit that sets importance groups of each application and each task to determine weights, and a power efficiency calculation unit that normalizes a task execution amount using the normalization coefficient, calculates a total work output by using the normalized task execution amount and the weight of the importance group of the application and the task, and calculates power efficiency.