Server Load Balancing via Physical Resource Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing systems for virtual machines do not effectively utilize availability information of physical resources to efficiently distribute server load, leading to suboptimal resource allocation and increased power consumption.
Innovation Solution
A system that accesses availability information of virtual machines running on physical machines, balances server load by selecting servers with loads below expansion thresholds and draining those below contraction thresholds, using a load balancer with an availability calculator, load distributor, and physical machine manager to dynamically adjust resource allocation and power modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If load balancing is performed without considering physical resource availability, then load distribution among virtual machines is simplified, but resource allocation efficiency deteriorates and power consumption increases
Solution Approach 1:
The patent implements nested monitoring where the load balancer monitors virtual machine performance metrics, which in turn monitor physical machine resource usage. This multi-layered monitoring structure allows the system to track physical resource availability through virtual machine performance data, improving resource allocation efficiency without requiring a completely separate monitoring system.
Solution Approach 2:
The patent introduces performance metrics as an intermediary that connects virtual machine operation status with physical resource availability. By using metrics such as CPU utilization, memory usage, and I/O performance as mediators, the load balancer can infer physical resource status without direct access, resolving the contradiction between simplicity and efficiency.
2Reliability
If more servers are kept active to handle load, then service availability is improved, but power consumption increases
Solution Approach 1:
The patent implements dynamic server state management where servers can transition between active, idle, and powered-off states based on real-time load conditions and performance metrics. The load balancer dynamically adjusts server availability by draining connections from underutilized servers and directing new connections to active servers, ensuring service availability while minimizing power consumption through state transitions.
Solution Approach 2:
The patent changes the operational parameters of servers based on performance metrics, adjusting CPU allocation, memory allocation, and power states according to actual resource utilization. By monitoring metrics and adjusting parameters dynamically, the system maintains service availability during peak loads while reducing power consumption during low-utilization periods.
3Productivity
If load is concentrated on fewer servers, then resource allocation efficiency is improved, but the risk of overload increases
Solution Approach 1:
The patent implements continuous feedback loops where the load balancer monitors performance metrics from servers and adjusts load distribution accordingly. When servers approach capacity thresholds or show signs of overload through performance degradation, the load balancer receives feedback and redistributes connections to prevent overload, maintaining both efficiency and reliability through adaptive feedback control.
Solution Approach 2:
The patent applies partial load concentration by directing a majority of connections to fewer optimized servers while maintaining standby capacity on additional servers. This allows efficient resource utilization on active servers while having backup capacity available to handle sudden load increases or failures, balancing efficiency gains with risk mitigation.
Data Source
Figure 1
Figure 2~3
AI summary
According to one embodiment, availability information describing virtual machines running on physical machines is accessed. The availability information associates each virtual machine with a physical resource used by the virtual machine. Use by the virtual machines is determined from the availability information. Availability of the physical resources is determined according to the use. Server load is balanced according to the availability of the physical resources. According to another embodiment, the following is performed until a load is accommodated: selecting a server having a load that is less than an expansion threshold; loading the server until the expansion threshold is reached; selecting a next server having a load that is less than a next expansion threshold; and loading the next server until the next expansion threshold is reached. Load of a low load server is determined to be below a contraction threshold, and the low load server is drained.