Dynamic Workload Adjustment for Data Center Power Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power consumption optimization methods for data centers fail to dynamically and continuously optimize power usage based on workload, as they only consider physical placement and peak temperature, neglecting device utilization.
Innovation Solution
A system and method that dynamically adjust workload distribution within a data center by generating candidate workload solutions, calculating temperature and performance profiles, and optimizing power consumption by selecting the solution with the lowest sum of power and migration costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional power consumption optimization methods are used that focus only on physical placement and peak temperature, then device placement is simplified, but power consumption cannot be dynamically optimized according to workload
Solution Approach 1:
The patent implements dynamic optimization by continuously monitoring device utilization metrics and workload characteristics, then adjusting workload distribution in real-time based on current conditions rather than relying on static placement decisions. This allows the system to adapt to changing workload patterns and optimize power consumption dynamically.
Solution Approach 2:
The system employs feedback mechanisms by monitoring device utilization, temperature, and power consumption metrics, then using this information to continuously adjust workload distribution. The feedback loop enables the system to learn from past performance and make informed decisions about optimal workload placement under varying conditions.
2Loss of energy
If device utilization is taken into account for power optimization, then power consumption can be optimized according to workload, but the optimization method becomes more complex
Solution Approach 1:
The patent optimizes power consumption by dynamically changing key parameters including device utilization levels, workload distribution patterns, and operational states based on monitored conditions. By adjusting these parameters in response to real-time data, the system achieves energy optimization without requiring overly complex algorithms.
3Loss of energy
If workload distribution is dynamically adjusted to reduce power consumption, then energy efficiency improves, but system performance may be affected
Solution Approach 1:
The system dynamically adjusts operational parameters such as device utilization and workload distribution to optimize energy efficiency while continuously monitoring performance metrics. By carefully controlling parameter changes and using feedback loops, the system maintains performance requirements while achieving energy savings.
Data Source
AI summary
A method and system for optimizing power consumption of a data center by dynamic workload adjustment. Workload of the data center is dynamically adjusted from a current workload distribution to an optimal workload solution. The optimal workload solution is a candidate workload solution of at least one candidate workload solution having a lowest sum of a respective power cost and a respective migration cost. Each candidate workload solution represents a respective application map that specifies a respective workload distribution among application programs of the data center. Dynamically adjusting the workload of the data center includes: estimating a respective overall cost of each candidate workload solution, selecting the optimal workload solution that has a lowest overall cost as determined from the estimating, and transferring the optimal workload solution to devices of a computer system for deployment.