Information Search System Using Named Memory and Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information search systems face challenges in efficiently searching large volumes of structured data, such as Terabyte-class information, due to high CPU load and memory requirements, leading to prolonged search times and inefficiencies in handling structured formats like XML or CSV.
Innovation Solution
An information search system utilizing a distributed network of computers, such as a cluster or array computer system, that allows parallel processing and uses named memories to store search results, eliminating the need for file read accesses by writing results directly to memory, thereby enhancing search speed and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trial and error information search is performed step by step on large-capacity data, then search completeness is improved, but search time becomes excessively long
Solution Approach 1:
The patent applies preliminary action by pre-processing and indexing information before the actual search operation. The information processing apparatus performs preliminary extraction of search targets and creates an index structure in advance, so that when a search is needed, the system can quickly retrieve results without performing complete trial-and-error searches on the entire large-capacity database, thus reducing search time while maintaining completeness.
2Reliability
If file read/write operations are performed for each search step, then search results can be stored persistently, but I/O time becomes significant and reduces efficiency
Solution Approach 1:
The patent applies dimensionality change by introducing a memory-based storage dimension alongside traditional file-based storage. The system uses memory (RAM) as an intermediate storage layer between the database and file system, allowing search results to be temporarily stored and accessed rapidly without immediate disk I/O operations. This adds a temporal dimension to data persistence, where data can be kept in memory for active use and only written to files when truly needed, thereby maintaining reliability while dramatically improving search efficiency.
3Adaptability or versatility
If CPU load is increased to handle structured information processing, then search capability is enhanced, but processing speed decreases due to memory constraints
Solution Approach 1:
The patent applies segmentation by dividing the information processing system into multiple independent information processing units that can operate in parallel. Each unit handles a portion of the structured information independently, allowing the system to process large volumes of data simultaneously without overwhelming the memory of a single unit. This segmentation enables enhanced search capability across distributed units while maintaining processing speed through parallel execution rather than sequential handling.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This invention provides an information search system including a management table which stores a name of a memory area and memory area information indicating the memory area in association with each other. When received source information search destination information indicates that source information to be searched is stored in a memory, the information search system searches the management table for a name of a memory area which matches a received source information name of the source information to be searched. When the name of the memory area which matches the received source information name of the source information to be searched is retrieved, the system acquires information which matches the received search target condition information from memory area information associated with the retrieved name of the memory area.