VM Grouping for Communication Speed and Heat Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data processing systems face inefficiencies due to applications and virtual machines being hosted on different servers, leading to reduced communication speed, increased heat generation, and unnecessary energy usage, especially during low processing demands.
Innovation Solution
A data processing system with a system controller that networks host servers to determine which applications or virtual machines communicate with each other, grouping them on the same server for improved efficiency and reducing server power usage by moving non-essential processes to offline mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If applications and virtual machines are hosted on different host servers, then system load is distributed evenly, but communication speed between applications decreases and heat generation increases
Solution Approach 1:
The patent groups applications and virtual machines that communicate frequently together on the same host server. This merging of related workloads eliminates the need for network communication between separate servers, thereby increasing communication speed and reducing heat generation from network infrastructure and server operations.
Solution Approach 2:
The system dynamically determines which applications and virtual machines should be grouped together based on their communication patterns. The controller continuously monitors and adjusts the placement of workloads on host servers, optimizing the balance between communication efficiency and heat generation in real-time.
2Productivity
If all servers in the cluster operate at full power, then processing capacity is maximized, but energy consumption increases unnecessarily during low processing demands
Solution Approach 1:
The system dynamically adjusts server power states based on actual processing demands. When processing load is low, the controller places certain host servers in offline mode to conserve energy. When demand increases, servers are brought online as needed, maintaining productivity while optimizing energy consumption.
Solution Approach 2:
Host servers are designed to be multi-functional, capable of operating in multiple states (online, offline, different power modes). This allows the same hardware infrastructure to serve both high-performance processing needs and energy-saving requirements, adapting to varying workload conditions.
3Productivity
If applications communicating with each other are hosted on the same host server, then communication efficiency improves, but server consolidation may increase load on individual servers
Solution Approach 1:
The controller dynamically balances workload distribution while maintaining communication efficiency. When grouping applications on the same server, the system monitors server load and can migrate workloads between servers to prevent any single server from becoming overloaded, thus maintaining both communication efficiency and acceptable power consumption levels.
Data Source
AI summary
A data processing system may include a plurality of host servers, each server including a device processor and a non-transitory computer readable medium including instructions stored thereon and executable by the processor, wherein the computer readable medium of each server includes instructions for running one or more virtual machines. In addition, the system may include a system controller including a device processor and a non-transitory computer readable medium including instructions stored thereon and executable by the processor, the computer readable medium of the system controller including instructions for receiving commands to host two or more virtual machines on the plurality of host servers; making a determination as to which of the virtual machines share commonalities with one another; and based on the determination, providing instructions to the plurality of host servers to host virtual machines determined to share one or more commonalities with one another on the same host server.