Least Hop Workload Transfer During Data Center Thermal Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data center disaster recovery methods are inefficient due to the inability to determine safe and available servers for workload transfer during thermal events, leading to potential server overload and application downtime.
Innovation Solution
A method that senses thermal events, identifies affected servers, and computes a weighted score for each server based on performance metrics, temperature, and proximity to the thermal event, to rank and select the optimal target server for workload transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If workload is transferred to servers farthest away from thermal event, then safety of transferred workload is improved, but target servers become overloaded and performance deteriorates
Solution Approach 1:
The patent changes the selection parameters from simple distance-based criteria to a multi-parameter scoring system that includes proximity to thermal event, current server performance metrics, and capacity availability. This allows dynamic optimization of target server selection based on real-time conditions, resolving the contradiction between safety and performance.
Solution Approach 2:
The system dynamically adjusts workload transfer decisions based on real-time sensing of thermal events, server performance metrics, and capacity conditions. Rather than static farthest-server selection, the system continuously evaluates and re-ranks potential target servers, enabling adaptive response to changing conditions that maintains both safety and performance.
2Reliability
If thermal event detection triggers immediate shutdown, then data center safety is improved, but application availability deteriorates
Solution Approach 1:
The system performs preliminary actions by detecting thermal events and automatically initiating workload transfer procedures before complete system shutdown becomes necessary. The multi-criteria ranking system identifies suitable target servers in advance, enabling seamless workload migration that maintains application availability while addressing the thermal safety issue.
Solution Approach 2:
The patent introduces an intermediary automated decision-making system that mediates between the thermal safety requirement and application availability. This system evaluates multiple factors including thermal proximity, server performance, and capacity to determine optimal workload relocation, avoiding both premature shutdown and unsafe transfer decisions.
3Measurement precision
If manual determination of safe servers is used, then transfer accuracy is improved, but response time deteriorates
Solution Approach 1:
The system implements self-service through automated sensing and evaluation of server conditions. Sensors continuously monitor thermal events and server performance metrics, and the multi-criteria ranking algorithm automatically determines safe target servers without manual intervention. This maintains high accuracy through systematic evaluation while dramatically reducing response time through automation.
Solution Approach 2:
The system incorporates feedback loops where sensors continuously monitor thermal conditions and server performance, feeding this information back to the workload transfer decision system. This real-time feedback enables dynamic re-evaluation of target servers and automatic adjustment of transfer decisions, maintaining precision while enabling rapid automated response.
Data Source
AI summary
An embodiment senses a thermal event in a data center, responsive to the sensed thermal event, identifies an affected server, and decides to migrate a workload from the affected server to a target server. The embodiment senses a performance metric, a temperature and a proximity from the thermal event for each of a plurality of servers, responsive to the sensed performance metric, the temperature and the proximity, computes a weighted score. The embodiment generates a ranking based on the weighted score of each of the plurality of servers. The embodiment selects the target server from the plurality of servers according to the ranking.