Search Query Skewing for Resource Load Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computing environments, scheduled search queries concentrate demand for computing and network resources at specific times, leading to potential data loss, system performance issues, and failure due to excess demand.
Innovation Solution
A method for skewing search queries over a period of time based on a search tolerance, distributing the demand for computing and network resources to reduce the likelihood of network data packet loss and failure modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If scheduled search queries are executed at fixed times, then search results can be obtained consistently, but computing and network resource demand concentrates at specific times causing data loss and system performance issues
Solution Approach 1:
The patent applies dynamics by making the search query execution times flexible rather than fixed. It introduces a skewing mechanism that randomly distributes query executions across a time window, allowing the system to adapt resource demand patterns dynamically while maintaining search functionality and reliability
Solution Approach 2:
The patent changes the time parameter of search query execution from fixed scheduled times to a distributed time window. By introducing a skewing factor that randomizes execution timing within a specified duration, it transforms the temporal parameter to distribute load while preserving search result consistency
2Object-affected harmful factors
If search queries are distributed over time, then resource demand concentration is reduced, but search result consistency may be affected
Solution Approach 1:
The system dynamically adjusts query execution timing within a skewing window while maintaining the same search criteria and data sources. This dynamic timing distribution reduces resource concentration while the consistent search logic ensures result reliability remains intact
Solution Approach 2:
The patent maintains continuous search functionality by ensuring that queries are executed within a defined time window rather than being completely redistributed. This continuous action approach preserves search result consistency while achieving load distribution through temporal spreading
3Loss of time
If multiple search queries are executed simultaneously, then data freshness is improved, but network data packet loss and system failures increase due to excess demand
Solution Approach 1:
The patent segments the simultaneous query executions into multiple individual queries distributed across a time window. Each query is executed separately at different times rather than simultaneously, reducing network and system load while maintaining acceptable data freshness through the skewing mechanism
Solution Approach 2:
The system implements periodic action by distributing query executions across regular time intervals within the skewing window. This periodic distribution prevents simultaneous executions that cause system overload while ensuring queries are executed frequently enough to maintain data freshness
Data Source
AI summary
Techniques for scheduling search queries in a computing environment are disclosed. A search query scheduling system associates a first set of queries with a first skew tolerance, the first set of queries scheduled to be performed during a first period, where the first skew tolerance is based on a duration of the first period. The search query scheduling system reschedules a first subset of search queries included in the first set of queries by skewing the first subset of search queries over a first portion of the first period based on the first skew tolerance.