Partial Query Result Merging in Multiserver Indexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multiserver computing environments, existing methods often send complete query results from multiple servers, resulting in excessive data transmission and processing, even when only a subset of results is needed, leading to inefficiencies in information retrieval.
Innovation Solution
The method involves distributing queries among partial index servers, calculating and merging only the required subset of results, and requesting N sorted results with an offset M from each server, where C (the number of results considered) is much greater than N, to efficiently generate the final merged result set, thereby reducing data sent over the network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all servers send complete partial query results to the dedicated server, then the complete result set can be obtained, but excessive data is transmitted over the network and most data is discarded
Solution Approach 1:
The patent extracts only the necessary portion of data from each server's complete result set. Instead of transmitting all partial results, the system identifies and sends only the top N results from each server to the dedicated server, eliminating the waste of transmitting data that will ultimately be discarded.
Solution Approach 2:
The patent applies partial action by sending slightly more data than strictly necessary (top N results from each server) rather than all available data. This partial transmission approach ensures the final sorted result set can be constructed without requiring complete data from all servers, achieving a balance between completeness and efficiency.
2Measurement precision
If all servers send complete partial query results, then accurate aggregated information can be collected, but query execution time increases due to merging large datasets
Solution Approach 1:
The patent extracts only the essential data elements needed for accurate aggregation - specifically the top N results from each server. This extraction maintains measurement precision for the final aggregated ranking while dramatically reducing the time required to collect and process data from multiple servers.
Solution Approach 2:
By transmitting partial results (top N from each server rather than complete sets), the system achieves sufficient accuracy for the final aggregated ranking. The partial data transmission is calibrated to provide exactly what is needed for accurate results without the time cost of processing complete datasets.
3Adaptability or versatility
If complete result sets are merged from multiple servers, then all possible results are available, but the majority of collected data is thrown away when only a subset is needed
Solution Approach 1:
The patent extracts only the necessary subset of data (top N results) from each server's complete result set. This extraction eliminates data waste by ensuring that only results that will be part of the final output are transmitted and processed, while still maintaining the ability to provide complete and accurate top N rankings.
Solution Approach 2:
The system transmits partial results from each server - enough to ensure the final top N can be accurately determined, but not so much that excessive data waste occurs. This partial transmission strategy achieves the optimal balance between data completeness and waste reduction.
Data Source
AI summary
A method and system for executing an information retrieval query in a multiserver computing environment is disclosed. The method and system employ a technique in which the query is distributed among each of a plurality of partial index servers in the multiserver environment, and a subset of results is calculated for each of the plurality of partial index servers. Then, the subset of results are merged in one logical index server to generate a merged result.