Distributed Node Scheduling with Multi-Dimensional Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing methods in distributed systems fail to effectively balance utilization of various resources across multiple dimensions, leading to limited overall resource utilization and high system costs.
Innovation Solution
A multi-dimensional load balancing method that determines a multi-dimensional load vector for each service node, classifies nodes into load intervals, and selects scheduling strategies to minimize the distance between the load vectors and a target load vector, ensuring balanced resource utilization across multiple dimensions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional load balancing methods (based on replicas, QPS, or disk space) are used, then single-dimension resource balance is achieved, but multi-dimensional resource utilization remains unbalanced
Solution Approach 1:
The patent transitions from single-dimension load balancing to multi-dimensional load balancing by constructing load vectors that encompass multiple resource dimensions (CPU, memory, disk, network, etc.). This dimensional expansion allows the system to simultaneously balance multiple resource types, resolving the contradiction between ease of single-dimension operation and multi-dimensional productivity.
Solution Approach 2:
The patent creates a composite load vector that integrates multiple resource metrics (CPU usage, memory usage, disk usage, network bandwidth, QPS, etc.) into a unified representation. This composite structure enables comprehensive resource evaluation and balancing across all dimensions, rather than treating each resource type in isolation.
2Productivity
If service replicas are concentrated on fewer nodes, then node utilization increases, but system cost and risk increase
Solution Approach 1:
The patent dynamically adjusts the distribution of service replicas across nodes based on multi-dimensional load parameters. By continuously monitoring and responding to changes in resource utilization metrics, the system optimizes replica placement to maintain high node utilization while avoiding concentration risks, thereby reducing system cost without compromising productivity.
3Quantity of substance
If service replicas are evenly distributed across nodes, then system cost decreases, but resource utilization efficiency is limited
Solution Approach 1:
The patent implements dynamic load balancing that continuously adapts replica distribution based on real-time multi-dimensional load conditions. This dynamic approach allows the system to achieve both cost efficiency and high resource utilization efficiency by flexibly adjusting allocations rather than maintaining static even distribution, resolving the contradiction between cost reduction and efficiency improvement.
Data Source
AI summary
The present disclosure relates to a load balancing method for a distributed system, an electronic device, and a non-transitory computer-readable storage medium. The method includes: determining a multi-dimensional load vector of each of service nodes based on dimensional resource usage indicators of each of service replicas carried on the each of the service nodes in the distributed system in multiple dimensions; classifying the service nodes into multiple load intervals based on the multi-dimensional load vector and a target load vector of each of the service nodes; determining multiple candidate scheduling strategies and corresponding load balancing gains based on the multiple load intervals; and selecting, from the multiple candidate scheduling strategies, a candidate scheduling strategy with a largest load balancing gain as a target scheduling strategy, and executing the target scheduling strategy.