Runtime Home Node Pairing for Data Store Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data processing systems have fixed relationships between home nodes and slave nodes, which limits flexibility and efficiency in managing memory access coherency and storage capacity, as they do not account for additional system properties such as memory container size, transmission paths, and link quality.
Innovation Solution
A method for runtime configuration that determines a modified size of the data store for each home node based on storage capacity and additional system properties, selecting a home node that satisfies a minimization criterion to pair with a slave node, optimizing the pairing by considering factors like memory capacity, transmission paths, and link quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed relationship between home nodes and slave nodes is used, then the system structure is simple and stable, but the flexibility and efficiency in managing memory access coherency and storage capacity are limited
Solution Approach 1:
The patent implements dynamic configuration of home node-slave node pairings by calculating modified data store sizes based on additional system properties (memory container size, transmission paths, link quality) and selecting optimal pairings at runtime. This allows the system to adapt to changing conditions without requiring a completely redesigned architecture, thus improving flexibility while managing complexity through algorithmic optimization.
Solution Approach 2:
The patent modifies the traditional fixed pairing parameters by introducing additional system properties as weighting factors. The modified data store size calculation incorporates memory container size, number of transmission paths, and link quality metrics, transforming the pairing decision from a static structural constraint to a dynamic optimization problem that can adapt to varying system conditions.
2Productivity
If conventional fixed pairing is used, then the configuration is simple, but system efficiency and capacity management are reduced
Solution Approach 1:
The patent employs feedback mechanisms by continuously monitoring system properties such as memory container size, transmission path characteristics, and link quality. This feedback information is used to recalculate modified data store sizes and optimize home node-slave node pairings dynamically, improving system efficiency through adaptive configuration rather than static setup.
Solution Approach 2:
The system performs self-optimization by automatically calculating modified data store sizes and selecting optimal home node pairings based on current system conditions. This self-service capability eliminates the need for manual configuration optimization while improving productivity through algorithm-driven decisions about resource allocation and pairing.
3Quantity of substance
If additional system properties are considered in pairing decisions, then optimized capacity utilization is achieved, but processing power and computation time increase
Solution Approach 1:
The patent applies partial action by considering only the most relevant system properties (memory container size, transmission paths, link quality) rather than all possible system parameters. This selective approach achieves sufficient capacity utilization optimization without the computational burden of analyzing every possible system attribute, balancing productivity improvement with acceptable processing requirements.
Data Source
AI summary
A data processing system and a method of runtime configuration of the data processing system are disclosed. The data processing system comprises a plurality of home nodes, and for a data store associated with a slave node in the data processing system, for each home node of the plurality of home nodes a modified size of the data store is determined. The modified size is based on a storage capacity of the data store and at least one additional property of the data processing system. A chosen home node of the plurality of home nodes is selected which satisfies a minimization criterion for the modified size, and the chosen home node is paired with the slave node.