Load Distribution Device Using Resource Consumption Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional server/client systems face inefficiencies in load distribution due to the lack of consideration for resource consumption patterns, leading to unbalanced resource usage and potential bottlenecks, resulting in delayed responses and reduced processing capacity.
Innovation Solution
A load distribution system that evaluates and selects servers based on their resource consumption patterns by calculating distances from an ideal consumption line, ensuring optimal resource allocation and balancing CPU and memory usage across servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If load distribution is performed using round-robin scheme or least connection scheme, then the load can be distributed among servers, but the distribution does not consider resource consumption patterns leading to unbalanced resource usage
Solution Approach 1:
The patent changes the load distribution parameters from simple metrics (number of sessions) to resource consumption metrics (CPU usage, memory usage). The determining unit calculates actual resource consumption for each server and uses these parameters to make distribution decisions, thereby achieving balanced resource utilization while maintaining productivity.
Solution Approach 2:
The system implements feedback by continuously monitoring resource consumption patterns of servers and using this information to dynamically adjust load distribution. The determining unit receives resource consumption data and adjusts server selection based on current resource usage states, creating a closed-loop control system that optimizes both productivity and resource balance.
2Device complexity
If servers are assigned based on number of sessions without considering resource retention, then load distribution is simplified, but resource sufficiency cannot be judged in advance causing response delays
Solution Approach 1:
The patent applies preliminary action by calculating and storing resource consumption patterns before actual load distribution occurs. The determining unit pre-assesses server resource capacity and consumption characteristics, enabling advance judgment of resource sufficiency. This allows the system to make informed distribution decisions without adding significant complexity, reducing response delays by having resource information ready beforehand.
3Ease of manufacture
If load distribution does not consider resource consumption, then implementation is simpler, but processes may be assigned to servers with insufficient resources leading to bottlenecks
Solution Approach 1:
The patent replaces the mechanical/simple approach of counting sessions with an information-based system that monitors and calculates resource consumption. Instead of using simple counters, the system uses determining units that calculate CPU and memory usage patterns, substituting a more complex measurement system for a simpler one to ensure resource sufficiency while maintaining reasonable implementation feasibility.
4Speed
If servers with unbalanced resource consumption are selected, then load distribution can proceed quickly, but processing capacity is reduced due to inability to execute new processes
Solution Approach 1:
The patent applies dynamics by making the load distribution system adaptive to changing resource consumption patterns. The determining unit continuously monitors server states and dynamically adjusts distribution decisions based on current resource availability. This dynamic approach ensures that load distribution remains fast while simultaneously optimizing processing capacity by selecting servers that can actually handle new processes.
Data Source
AI summary
In a server/client system in which a plurality of servers and a client are connected through a network, and process-executing servers execute a process based on a process request from the client and transmit a process result to the client, a load distribution device includes a process information receiving unit that receives information on the process from the client; a determining unit that determines a server to execute the process from among the process-executing servers based on the information on the process; and a server information transmitting unit that transmits information on determined servers to the client, and the client includes a server information receiving unit that receives transmitted information on the servers; and a process request transmitting unit that transmits the process request to the determined servers.