State Chart Execution Device Optimizing Memory Layout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional state chart execution methods are inefficient for large-scale state charts due to suboptimal data structure organization in memory, leading to decreased execution speed, especially when dealing with hierarchical and parallel states.
Innovation Solution
A state chart execution device that converts state chart information into binary data by calculating relevance between data elements, binarizing them, and arranging them in an optimal array with set offsets to improve memory access efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a dynamic tree structure is used to store state chart data, then the state chart can be flexibly organized, but the cache hit ratio decreases and processing time increases when the number of states becomes enormous
Solution Approach 1:
The patent changes the data structure from a dynamic tree structure to a binary data structure with optimized array arrangement. By calculating relevance between state data elements and arranging them in arrays based on this relevance, the system achieves both flexibility and high performance, resolving the contradiction between adaptability and productivity.
2Ease of operation
If state chart data is stored in a table format with transition history, then the state chart can be executed systematically, but the execution speed lowers when the state chart has a hierarchical structure and an enormous number of states
Solution Approach 1:
The patent transforms the table-based data structure into a binary data structure where state data elements are arranged in arrays according to their relevance. This parameter change maintains systematic execution capability while dramatically improving execution speed by optimizing memory layout and cache utilization.
3Device complexity
If related state data is arranged at remote positions in memory, then the data structure can be simple, but the cache hit ratio decreases and processing becomes slow
Solution Approach 1:
The patent performs preliminary arrangement of state data elements in arrays based on calculated relevance before execution. This preliminary action ensures that related data is positioned close together in memory, improving cache hit ratio and processing speed without adding complexity during execution.
Solution Approach 2:
The patent changes the memory arrangement parameter from random or simple structure to optimized array arrangement based on relevance. This parameter change achieves both simplicity and high performance by pre-organizing data according to access patterns.
Data Source
AI summary
A relevance calculation unit calculates relevance between pieces of data constituting a state chart. An individual data binarization unit binarizes the pieces of data constituting the state chart. An optimal array calculation unit calculates an array of the pieces of data constituting the state chart on the basis of the relevance calculated by the relevance calculation unit. An offset setting unit sets an offset indicating a position of each of the pieces of data binarized by the individual data binarization unit in a case where the binarized pieces of data are arranged in the array calculated by the optimal array calculation unit. A binary data integration unit integrates the pieces of data each of which has been binarized by the individual data binarization unit and for each of which the corresponding offset has been set by the offset setting unit in the array calculated by the optimal array calculation unit to obtain binary data. A state chart execution unit executes processing of the state chart on the basis of the binary data.