Elastic Database Pool Requisition via Predictive Multiplexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing systems face inefficiencies due to underprovisioned or overprovisioned database-compute tiers, leading to performance degradation and inefficient resource deployment, as existing methods fail to accurately account for multiple activity levels and peak utilization patterns across databases.
Innovation Solution
A prescriptive analytics-based multi-layered elastic requisition stack that uses historical and predicted data to identify candidate databases for multiplexing into shared resource pools, optimizing database-compute tier selection and resource allocation through deep-learning analysis and statistical multiplexing, ensuring accurate forecasting and efficient resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional database-compute tier requisition methods are used, then resource allocation is simplified, but accuracy of provisioning deteriorates leading to underprovisioning or overprovisioning
Solution Approach 1:
The patent segments the requisition system into multiple layers including data input layer, candidate layer, pool layer, requisition layer, consumption savings layer, and presentation layer. Each layer processes specific aspects of resource allocation independently, improving provisioning accuracy through specialized analysis while managing complexity through modular architecture.
Solution Approach 2:
The system performs preliminary actions by identifying candidate databases for multiplexing before final requisition decisions. The candidate layer pre-processes database characteristics and utilization patterns to prepare optimized pool configurations, ensuring accurate provisioning before resources are actually allocated.
2Productivity
If database-compute resources are statically allocated, then system configuration is simplified, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent implements dynamic resource allocation where the requisition system continuously monitors database utilization patterns and adjusts pool configurations accordingly. The system dynamically identifies candidate databases for multiplexing based on real-time performance metrics, enabling flexible resource allocation that adapts to changing workload demands.
Solution Approach 2:
The system incorporates feedback mechanisms where consumption metrics and utilization patterns are continuously monitored and fed back into the candidate layer analysis. This feedback loop enables the system to learn from past allocation decisions and optimize future resource provisioning, improving utilization efficiency while managing complexity through data-driven adjustments.
3Adaptability or versatility
If multiple databases are multiplexed into shared pools, then resource sharing is improved, but performance degradation risk increases
Solution Approach 1:
The patent applies local quality by analyzing and grouping databases with similar utilization patterns and performance characteristics into specific pools. The candidate layer evaluates individual database attributes locally to determine compatibility for multiplexing, ensuring that databases with complementary peak utilization profiles are grouped together while maintaining performance stability.
Solution Approach 2:
The system changes parameters by selecting databases based on specific utilization patterns, peak timing, and performance metrics before multiplexing. The requisition layer adjusts pool configurations based on analyzed parameters, optimizing resource sharing capability while maintaining performance stability through parameter-based selection criteria.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A multi-layer elastic requisition stack may generate pool requisition tokens for controlling requisition of pooled database-compute resources. The elastic requisition stack may determine candidate databases for inclusion in elastic pools by analyzing historical utilization data and generating predicted utilization data. Based on the historical and predicted utilization data, the elastic requisition stack may determine multiplexing characteristics for the candidate databases and complement factors among the databases. The elastic requisition stack may compare unpooled database performance to pooled database performance to determine whether to pool the candidate databases.