Multi-Slice Indexing for Rapid Content Serving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing indexing systems for electronic content are inefficient due to their specificity to content type, allowing only simultaneous reading or writing, leading to delayed searches and maintenance challenges, and are slow in processing textual content, which hinders the ability to quickly serve the most valuable content.
Innovation Solution
A multi-slice indexing system where each slice contains multiple index entries, with updates queued and collapsed into single operations, allowing all slices to remain searchable during maintenance, using integer tokens to represent attributes and properties for rapid comparison and ranking of content items, and optimizing entry ordering for quick identification of the most valuable content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a specialized index is used for one type of content, then the indexing is optimized for that content type, but the index cannot be used effectively with other types of content
Solution Approach 1:
The patent implements a universal index structure that can handle multiple content types (web pages, advertisements, documents, news stories) through a common entry format containing identifiers, titles, textual content, and metadata. This allows a single index to serve multiple content types effectively, eliminating the need for separate specialized indexes for each content type while maintaining optimization for each type through structured fields.
2Ease of manufacture
If the index is locked for writing during maintenance, then updates can be applied, but the index cannot be used to locate and serve content during maintenance
Solution Approach 1:
The patent divides the index into multiple segments or partitions that can be maintained independently. During maintenance, only specific segments are locked for writing while other segments remain accessible for reading and searching. This segmentation allows concurrent read and write operations on different parts of the index, reducing search delays during maintenance cycles.
Solution Approach 2:
The patent implements preliminary actions by creating backup copies of index segments before maintenance operations. The index structure allows readers to continue accessing the current version while writers prepare updated versions in advance. This preliminary preparation enables seamless switching between versions without interrupting search operations.
3Ease of manufacture
If multiple updates are applied in serial fashion during maintenance, then all updates are processed, but the maintenance duration increases proportionally
Solution Approach 1:
The patent merges multiple update operations into consolidated maintenance cycles. Instead of applying updates serially one at a time, the system batches multiple updates and applies them together in optimized maintenance operations. The index structure supports merging of duplicate entries, updates to existing entries, and batch insertions, significantly reducing maintenance duration compared to serial processing.
4Measurement precision
If textual content is read, parsed and compared with query parameters, then accurate search results are obtained, but the scanning speed decreases
Solution Approach 1:
The patent applies preliminary action by pre-processing textual content during indexing to extract and store key features, keywords, and metadata in structured formats. This preliminary extraction allows queries to compare against pre-computed features rather than parsing full textual content during search, significantly improving scan speed while maintaining search accuracy through preserved semantic information.
Solution Approach 2:
The patent extracts essential semantic information from textual content during the indexing phase and stores it in separate, efficiently queryable fields such as keywords, tags, and metadata. This extraction separates the heavy textual parsing from the query operation, allowing rapid comparison of extracted features while maintaining the ability to retrieve accurate search results based on the original content.
Data Source
AI summary
Apparatus and methods are provided for indexing electronic content to be served to users' mobile and/or stationary communications and computing devices. An index is composed of multiple slices, with each slice storing multiple entries and each entry representing one content item or one campaign or collection of content items. An entry is populated with tokens representing attribute/value pairs of a target audience of the content item and/or property/value pairs of the item or the item's campaign. A query or request to identify content items for serving to a particular user is similarly formatted with tokens representing attribute/value pairs of the user and/or item/campaign. Queries can then be executed rapidly across any or all index entries in any or all slices. Within a slice, entries may be sorted by value or score, and integer components within an individual entry may be sorted to facilitate rapid comparison with a query.