Server Farm Energy Efficiency via Virtual Server Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Server farms face inefficiencies in energy usage due to uneven distribution of computational tasks across servers with varying energy efficiency, leading to suboptimal energy efficiency and increased greenhouse gas emissions.
Innovation Solution
A method and system for operating a server farm that sorts servers by energy efficiency, forms virtual servers from the most efficient ones, and assigns tasks to maximize energy efficiency by prioritizing servers with vacant slots and optimal service rates, allowing for dynamic adjustment based on task arrival rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computational tasks are distributed across all servers in the server farm, then job throughput is maintained, but energy efficiency deteriorates due to uneven distribution across servers with varying energy efficiency
Solution Approach 1:
The patent applies local quality by creating virtual servers that group physical servers with similar energy efficiency characteristics. Jobs are routed to specific virtual servers based on energy efficiency requirements, ensuring that each job is processed by servers with appropriate energy characteristics rather than distributing jobs uniformly across all servers. This localized assignment optimizes energy efficiency while maintaining throughput.
Solution Approach 2:
The patent segments the server farm into multiple virtual servers, each comprising a subset of physical servers with comparable energy efficiency. This segmentation allows the system to selectively engage energy-efficient servers for specific job types while keeping energy-intensive servers available for other purposes, thereby improving overall energy efficiency without sacrificing total job throughput.
2Loss of energy
If servers are consolidated to improve energy efficiency, then power consumption is reduced, but system complexity increases due to virtual server management
Solution Approach 1:
The patent introduces a job assignment policy as an intermediary layer between jobs and physical servers. This policy automatically routes jobs to appropriate virtual servers based on energy efficiency criteria, eliminating the need for manual server management while achieving energy consolidation benefits. The intermediary handles the complexity of virtual server management transparently.
Solution Approach 2:
The system implements self-service through automated job assignment policies that dynamically route jobs to energy-efficient servers without human intervention. The virtual server architecture automatically manages server consolidation and distribution, reducing operational complexity while maximizing energy efficiency gains from server consolidation.
3Use of energy by moving object
If energy-efficient servers are prioritized for job assignment, then energy efficiency improves, but adaptability decreases when job arrival rates fluctuate
Solution Approach 1:
The patent implements dynamics by making the job assignment policy adaptive to changing job arrival rates. When job arrival rates increase, the system can dynamically adjust by engaging additional servers including those with lower energy efficiency, ensuring throughput is maintained. When arrival rates decrease, the system returns to prioritizing energy-efficient servers, maintaining optimal energy efficiency. This dynamic adjustment resolves the contradiction between energy efficiency and adaptability.
Data Source
AI summary
A method for operating a server farm with a plurality of servers operably connected with each other includes the steps of: sorting the plurality of servers according to a respective energy efficiency value associated with each of the plurality of servers; defining a virtual server by selectively joining two or more of the servers with the highest energy efficiency values; receiving a job request of a computational task to be handled by the server farm; and assigning the computational task to one of the plurality of servers for processing based on a job assignment policy, the job assignment policy being arranged to give preference to assigning the computational task to the servers of the virtual server for processing so as to maximize an energy efficiency of the server farm.