Memory Searching System Using Pointer Switching for Real-Time Index Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing real-time search systems face high costs and low efficiency due to the integration of searching and indexing processes, which consume significant computational resources, especially when dealing with large datasets.
Innovation Solution
A memory searching system and method that separates indexing and searching modules, utilizing multiple indexed databases for incremental synchronization and real-time updating, allowing for efficient data management and reduced hardware costs by switching pointers in memory databases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If indexing and searching are merged into one process, then real-time search capability is achieved, but computational resource consumption increases significantly
Solution Approach 1:
The patent divides the search system into two independent modules: an indexing module that processes and indexes data, and a searching module that performs search operations. This segmentation allows the searching module to use pre-built indexes without participating in the indexing process, thereby reducing its computational resource consumption while maintaining real-time search capability.
Solution Approach 2:
The indexing module performs preliminary indexing of data before search operations are needed. By pre-processing and building indexes in advance, the searching module can directly utilize these prepared indexes for fast retrieval, avoiding the need to perform indexing operations during search time and thus reducing computational resource consumption.
2Speed
If indexing and searching are integrated in one process, then real-time search is supported, but hardware equipment requirements increase
Solution Approach 1:
The system architecture is segmented into separate indexing and searching modules that can be deployed independently. The searching module only requires memory resources to store and access indexes, while the indexing module handles data processing. This segmentation reduces the hardware requirements for the searching module, eliminating the need for it to have the full computational capabilities required for both indexing and searching.
3Quantity of substance
If indexed data in memory reach a threshold, then data is merged into disk, but search efficiency decreases due to dual-location searching
Solution Approach 1:
The patent extracts the indexing function from the searching module and places it entirely in the indexing module. The searching module only performs search operations on data that is already indexed in memory, without needing to search in disk locations. This extraction eliminates the need to search both memory and disk simultaneously, thereby maintaining search efficiency even when large volumes of data are managed.
Data Source
AI summary
Described is a memory searching system, including an indexing module and a searching module. The indexing module, provided with multiple indexed databases, acquires and records updated indexed data in an indexed database provided to the indexing module and synchronizes the searching module with the updated indexed data recorded in the indexed database provided to the indexing module. The searching module updates an indexed database in a memory with the updated indexed data received from the indexing module (the indexed database in the memory corresponds to the indexed database provided to the indexing module), and switches a pointer of the indexed database to the updated indexed database in the memory. Also described is a memory searching method, a real-time searching system and method, capable of better supporting real-time search and of updating indexed data in seconds in a realistic application, reducing the cost of hardware equipment and improving searching efficiency.