Load Sharing System with Distributed Cache Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication systems face challenges in managing load sharing and cache management across multiple nodes, as single-node systems can no longer meet increasing communication demands.
Innovation Solution
A system and method for load sharing that includes a distribution device, service processors, and inquiry processors, where the distribution device assigns service processing requests to target service processors, which then determine and request data from target inquiry processors. If data is not available, the service processor accesses the core network, and stores data from the core network to the inquiry processor's memory, excluding inaccessible inquiry processors and automatically deleting expired data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single-node processing system is used, then the system structure is simple, but it cannot meet the increasing communication demand
Solution Approach 1:
The system is divided into multiple independent nodes including service processors and inquiry processors, each capable of handling specific tasks. This segmentation allows the system to scale horizontally to meet increasing communication demands while maintaining manageable complexity through modular architecture.
Solution Approach 2:
Multiple service processors and inquiry processors are combined into a coordinated multi-node system. The service processors handle service processing requests while inquiry processors manage data inquiries, with both types of processors working together to achieve higher overall productivity than a single-node system.
2Productivity
If multiple nodes are introduced for load sharing, then the processing capacity increases, but the cache management complexity increases
Solution Approach 1:
Each inquiry processor maintains its own local cache memory and autonomously manages its cached data. When a service processor needs data, it queries the inquiry processor's cache first before accessing the core network. This self-service approach distributes cache management responsibilities across nodes, reducing overall management complexity while maintaining high processing capacity.
Solution Approach 2:
Inquiry processors pre-fetch and cache data from the core network before it is needed by service processors. This preliminary action reduces the burden on the core network and allows service processors to retrieve data quickly from local caches, thereby increasing processing capacity without proportionally increasing cache management complexity.
3Speed
If data is cached in inquiry processors, then data access speed improves, but data freshness may deteriorate
Solution Approach 1:
The system implements a time-based cache invalidation mechanism where cached data is periodically refreshed or invalidated based on preset time periods. This periodic action ensures that data in the cache remains relatively fresh while still benefiting from cached access speeds, balancing both data access speed and data freshness requirements.
Solution Approach 2:
The system employs a query-based feedback mechanism where service processors query inquiry processors for data availability. When data is not found in the cache or is stale, the system retrieves fresh data from the core network and updates the cache accordingly. This feedback loop maintains data freshness while preserving the speed benefits of caching for valid cached data.
Data Source
Figure 1~2
Figure 3
AI summary
Disclosed are a system and a method for load sharing, and a computer storage medium. The system includes a distribution device, a plurality of service processors and a plurality of inquiry processors. A target service processor is determined from the service processors by the distribution device; the distribution device sends a service processing request to the target service processor, so that the target service processor can determine a target inquiry processor from the inquiry processors, and further send a target data inquiry request to the target inquiry processor, and the target inquiry processor can send target data to the target service processor after acquiring the target data cached in an own memory; after acquiring the target data, the target service processor carries out a service work needed to be applied to the target data.