Dynamic Weight Allocation for Cloud Resource Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems face challenges in efficiently allocating resources across multiple clients with varying usage patterns and data sizes, leading to inefficiencies and limitations in performance scaling.
Innovation Solution
A computer-implemented method that dynamically assigns weights to client systems based on their data usage and category, prioritizing resource allocation for data operations, and balancing weights across node sets to accommodate potential growth and optimize resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If weights are assigned to client systems to maintain performance level, then service level agreement is met, but client system growth is limited by required performance level
Solution Approach 1:
The patent implements dynamic weight adjustment where client system weights are continuously monitored and updated based on actual resource consumption patterns. The system transitions from static performance-level weights to dynamic usage-based weights, allowing clients to grow their resource allocation as their actual usage increases, thereby resolving the contradiction between maintaining performance levels and allowing client growth.
Solution Approach 2:
The system changes the parameter basis for weight assignment from fixed performance metrics to variable usage parameters. By monitoring actual resource consumption (storage, processing, network bandwidth) and adjusting weights based on these changing parameters, the system enables clients to scale their resource allocation dynamically rather than being capped by predetermined performance thresholds.
2Reliability
If weights are adjusted to meet performance metrics, then service level agreement is satisfied, but resource distribution efficiency deteriorates
Solution Approach 1:
The system implements self-service resource allocation where clients automatically receive resource allocation proportional to their actual usage patterns. Instead of manual weight adjustment to meet SLAs, the system autonomously monitors usage and distributes resources based on real-time consumption, eliminating the inefficiency of forced distribution while maintaining performance commitments.
Solution Approach 2:
The patent implements continuous feedback loops where resource consumption is monitored, weights are adjusted based on actual usage patterns, and allocation is reallocated accordingly. This feedback mechanism ensures that resources are distributed efficiently based on actual needs rather than predetermined metrics, resolving the contradiction between meeting performance goals and maintaining distribution efficiency.
3Productivity
If different weights are assigned to client systems for various nodes, then resource allocation is optimized, but client system stability deteriorates due to frequent node switching
Solution Approach 1:
The system segments resource allocation at the node level while maintaining client-level weight consistency. Each node independently applies the client's weight to determine its share of resources, allowing optimized local allocation without requiring client reassignment between nodes. This segmentation enables resource optimization while preserving client stability across the distributed system.
Data Source
AI summary
A computer-implemented method may include availing a computer platform to a client system. The computer platform may host data and perform data processing. The computer platform may receive a data set from the client system. Computational resources may be determined for the client system, based at least in part on the size of the data set and a size of other data received from the client system, and a category assigned to the client system. A weight may be determined for the client system based on the category, indicating a degree to which resources may be preferentially allocated to process requests associated with the client system. The client system may request a data operation be performed using the data set. Allocation of one or more resources may be prioritized for the data operation, based on the weight. The prioritization may influence how quickly the data operation is performed.


