Cloud Controller Job Migration for Anticipated Power Outages
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face challenges in automatically migrating computing jobs during anticipated power outages, relying on manual human analysis and processes that are inefficient and may not account for real-time power system data, especially in emergency situations where administrators may be unreachable.
Innovation Solution
A method that receives power-related data to determine anticipated power outages and automatically identifies jobs to migrate from a primary data center to a remote one, using machine logic and historical data to prioritize and manage the migration process, ensuring seamless disaster recovery and avoiding power outages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual human analysis is used to decide job migration during power outages, then administrators can make informed decisions, but the process is inefficient and may not account for real-time power system data
Solution Approach 1:
The system enables autonomous self-service by implementing automated monitoring and decision-making capabilities that allow the data center to independently analyze power system data and execute job migration decisions without human intervention, resolving the contradiction between manual analysis accuracy and automated efficiency
Solution Approach 2:
The patent replaces the mechanical human decision-making process with an automated electronic system that continuously monitors power system data and automatically executes migration decisions based on pre-defined criteria, eliminating the inefficiency of manual processes while maintaining decision quality through sophisticated algorithms
2Productivity
If automated migration is implemented, then decision-making efficiency is improved, but the system lacks autonomous capability to make decisions without human input
Solution Approach 1:
The system applies preliminary action by pre-configuring migration policies, thresholds, and decision-making rules before power outages occur. The automated system is pre-programmed with the authority and logic to make autonomous decisions when specific conditions are met, eliminating the need for real-time human approval while maintaining controlled automation
Solution Approach 2:
The patent implements continuous feedback loops where the system monitors power system status, compares readings against pre-defined thresholds, and automatically adjusts migration decisions based on real-time conditions. This feedback mechanism enables true autonomous operation by allowing the system to self-correct and adapt without human intervention
3Measurement precision
If real-time power system data is continuously monitored, then outage prediction accuracy is improved, but the system complexity increases
Solution Approach 1:
The patent applies universality by designing a multi-functional monitoring system that simultaneously performs data collection, analysis, prediction, and decision-making. This integrated approach consolidates multiple functions into a unified system, reducing overall complexity compared to separate specialized systems while maintaining high prediction accuracy through comprehensive data processing
Data Source
AI summary
An approach is disclosed that receives power related data from one or more power systems. The approach then determines, based on an analysis of the power related data, an anticipated power outage, with the power outage includes a power outage time estimate. The approach further identifies jobs to be migrated from a primary data center to a remote data center. The identification of the jobs to be migrated is based, at least in part, on the power outage time estimate.