Low-Load Data Group Query Routing Across Execution Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face increased processing load and performance deterioration during scale-out operations due to the addition of new containers with large table constellations, leading to service disruptions and elevated operational costs.
Innovation Solution
A control method that identifies low-processing-load data groups and redistributes queries to alternative execution environments, reducing the need for new container additions and minimizing load on the system by optimizing resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If new containers with large table constellations are added during scale-out operations, then system capacity and processing capability are improved, but processing load increases and system performance deteriorates
Solution Approach 1:
The patent extracts and identifies specific data groups with low processing load from the existing container using statistical analysis. By separating these low-load data groups from the main container, the system can redistribute them to alternative execution environments, thereby reducing the processing load on the original container during scale-out operations while maintaining overall system capacity.
Solution Approach 2:
The patent segments the data constellation into different data groups based on processing load characteristics. By dividing the large table constellation into manageable segments with different load profiles, the system can selectively migrate only the low-load segments to new containers, minimizing the impact on system performance during scale-out while still achieving capacity expansion.
2Productivity
If new containers are added to handle increased load, then processing capability is improved, but operational costs increase
Solution Approach 1:
The patent implements dynamic query routing that automatically directs queries to the most appropriate execution environment based on real-time load conditions. This dynamic approach allows the system to maximize the utilization of existing containers before adding new ones, thereby improving processing capability while minimizing the need for additional container additions and reducing operational costs.
Solution Approach 2:
The patent creates a multi-functional system where existing containers can serve multiple purposes: handling their original data groups while also receiving redistributed low-load data groups from other containers. This universal utilization of existing resources delays the need for new container additions, improving processing capability without proportionally increasing operational costs.
3Reliability
If queries are redistributed to existing execution environments, then processing load increase is suppressed, but system complexity increases
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors processing load statistics for each data group and automatically adjusts query routing decisions based on this feedback. The statistical information about processing loads guides the redistribution process, allowing the system to suppress processing load increases through automated, data-driven decisions rather than complex manual configuration, thereby managing system complexity through intelligent automation.
Data Source
AI summary
A control method in which a computer executes processing includes: specifying a first data group of which a processing load of a query is less than a first reference value, in a first data constellation, based on statistical information that indicates a statistical value regarding the processing load of the query, for each data group to be an access unit of the query, in the first data constellation included in a first execution environment included in a plurality of execution environments that has different data constellations; specifying a second execution environment that has a second data constellation that includes the specified first data group, included in the plurality of execution environments, other than the first execution environment; and controlling the plurality of execution environments to transmit a new query that uses the specified first data group as an access unit to the specified second execution environment.