Renewable Energy Scheduling in Computer Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scheduling tasks in a computer cluster is complicated by the need to consider multiple parameters, including computing resources, deadlines, and the variability of renewable electricity sources, which introduces additional constraints and inefficiencies due to the failure risk of these sources.
Innovation Solution
A method that predicts and manages the failure risk of renewable electricity sources by adjusting their availability and reconfiguring them dynamically based on real-time need and availability predictions, incorporating feedback from scheduling performances to improve prediction accuracy and reduce disturbance in the scheduling process, while also considering electricity price and workload to optimize energy use.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the failure risk of renewable electricity source is taken into account as a parameter in scheduling, then the reliability of electricity supply is improved, but the complexity of the scheduling algorithm increases
Solution Approach 1:
The patent applies preliminary action by predicting the failure risk of renewable electricity sources in advance before scheduling tasks. The system performs prediction of failure risk, prediction of electricity need, and prediction of electricity availability beforehand, then uses these pre-computed predictions to guide scheduling decisions. This resolves the contradiction by preparing reliability data ahead of time, allowing the scheduling algorithm to use reliable predictions without adding real-time computational complexity.
Solution Approach 2:
The patent implements feedback by using the predicted failure risk, electricity need, and electricity availability as input parameters that continuously guide and adjust the scheduling algorithm. The scheduling decisions are made based on feedback from these predictions, creating a closed-loop system where reliability information flows into the scheduling process. This resolves the contradiction by systematically integrating reliability feedback without requiring complex real-time adjustments.
2Measurement precision
If the scheduling algorithm considers multiple parameters including failure risk, then the quality of scheduling decisions is improved, but the speed of making scheduling decisions deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing predictions of failure risk, electricity need, and electricity availability before the scheduling decision is required. These predictions are prepared in advance and then used as input parameters for the scheduling algorithm. This resolves the contradiction by separating the computationally intensive prediction processes from the time-critical scheduling decision, allowing high-quality multi-parameter considerations without sacrificing decision speed.
3Loss of energy
If renewable electricity sources are used to lower global energy consumption cost, then the energy cost is reduced, but the scheduling efficiency deteriorates due to additional constraints
Solution Approach 1:
The patent applies parameter changes by transforming the scheduling problem from a traditional resource-allocation task into a multi-parameter optimization problem that includes failure risk, electricity need, and electricity availability. By changing the parameters considered in scheduling (adding reliability and renewable-specific parameters), the system can optimize for energy cost while maintaining efficiency through structured prediction-based decision making rather than unstructured constraint handling.
Solution Approach 2:
The patent applies preliminary action by pre-computing the failure risk, electricity need, and electricity availability parameters before scheduling. This allows the scheduling algorithm to work with pre-prepared data, reducing the computational burden during the actual scheduling process. This resolves the contradiction by enabling the system to consider multiple renewable-related parameters without proportionally increasing scheduling complexity, thus maintaining efficiency while reducing energy costs.
Data Source
Figure 1
AI summary
This invention relates to a method of managing electricity providing in a computers cluster (9), comprising: a process of prediction of need of electricity (8) provided by at least one renewable electricity source (1, 2, 3) in said computers cluster (9), a process of prediction of availability of said electricity (6) provided by said renewable electricity source (1, 2, 3), including: a step of managing failure risk of said renewable electricity source (1, 2, 3), by lowering said predicted availability, so as to: increase life expectancy of said renewable electricity source (1, 2, 3), and/or lower maintenance frequency of said renewable electricity source (1, 2, 3), a process of scheduling tasks (7) in said computers cluster (9), based on both said prediction processes (6, 8).