Faceted Search Multipath Indexing for Lower Query Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search technologies require traversal of all stored content to access documents, leading to inefficiencies and increased latency in processing search requests.
Innovation Solution
A multipath index is generated and populated with subpath index entries, allowing direct access to document content through conditional and facet subpaths, reducing the need to retrieve entire documents from storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional search methods are used to access stored content, then direct access to documents is achieved, but traversal of all stored content is required leading to increased latency
Solution Approach 1:
The patent segments the stored content into multiple independent indexes organized in a tree structure with hierarchical levels. Each index contains a subset of the total content, allowing search requests to be processed by traversing only relevant segments rather than the entire content store, thereby reducing latency and improving access speed.
2Productivity
If entire documents are retrieved from storage to service search requests, then complete document access is achieved, but I/O operations increase leading to reduced efficiency
Solution Approach 1:
The patent extracts and stores only the necessary content portions in the hierarchical indexes rather than retrieving entire documents. The index structure captures essential metadata and content fragments that can satisfy search requests without requiring full document retrieval, significantly reducing I/O operations and improving processing efficiency.
Solution Approach 2:
The patent performs preliminary indexing of content into a hierarchical structure before search operations. This pre-processing organizes content into manageable segments with metadata that enables rapid search request evaluation without requiring full document access, thereby reducing I/O overhead and improving efficiency.
3Ease of operation
If a simple index structure is used for direct document access, then ease of access is achieved, but inability to efficiently handle faceted search and conditional subpaths occurs
Solution Approach 1:
The patent implements a universal hierarchical index structure that can handle multiple search types including direct document access, faceted search, and conditional subpath queries. The tree-organized indexes with configurable levels and metadata provide a flexible framework that adapts to various search requirements without sacrificing direct access capabilities.
Data Source
AI summary
In general, embodiments of the technology relate to a method for multipath searching. The method includes generating a multipath index and using the multipath index to service a request. Generating the multipath index includes obtaining a main path, a first subpath and a plurality of additional subpaths from a multipath index definition, identifying documents that include the first subpath, obtaining values for the first subpath and the plurality of additional subpaths from the documents, creating subpath index entries for the documents that include the values for the additional subpaths, and populating the multipath index with the created subpath index entries. The subpath index entries in the multipath index can be associated with keys corresponding to values associated with the first subpath.