Priority-Based Cloud Bandwidth Flow Control for Oversold Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud service providers oversell bandwidth resources, leading to performance bottlenecks and degradation when multiple users access simultaneously, affecting all users and potentially causing crashes.
Innovation Solution
Implement a flow control method that assigns different priorities to user sets and periodically reallocates bandwidth resources based on previous allocation cycles, calculating allocable bandwidth resources for each user set and user within those sets to manage access requests dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If bandwidth resources are oversold to multiple users, then the quantity of bandwidth resources sold increases, but the physical bandwidth resources reach a performance bottleneck causing performance degradation
Solution Approach 1:
The system segments users into different priority groups (first priority user set, second priority user set, etc.) and allocates bandwidth resources to each segment separately. This segmentation allows the system to manage oversold bandwidth by ensuring that high-priority users receive guaranteed resources while lower-priority users can utilize remaining capacity, thus preventing performance degradation across all users simultaneously.
Solution Approach 2:
The system dynamically adjusts the allocation parameters of bandwidth resources based on actual usage and priority levels. By changing the allocation ratio and available bandwidth parameters for different user sets, the system can accommodate more users than physical resources would normally support while maintaining service quality for critical users, effectively resolving the contradiction between overselling and performance reliability.
2Productivity
If multiple users access based on maximum bandwidth resources simultaneously, then user access capacity increases, but the physical bandwidth resources reach a bottleneck causing all users to be affected
Solution Approach 1:
The system applies different quality levels of service to different user groups. High-priority users receive guaranteed bandwidth resources with higher quality assurance, while lower-priority users receive remaining resources. This local differentiation of service quality allows the system to maintain high access capacity for critical users while still permitting lower-priority access, preventing universal performance degradation.
Solution Approach 2:
The system implements dynamic bandwidth allocation where the available bandwidth for each user set is adjusted in real-time based on current system load and user priorities. When physical resources are under pressure, the system dynamically reduces allocation to lower-priority users while maintaining service for high-priority users, thus managing access capacity without causing widespread performance degradation.
3Device complexity
If bandwidth resources are allocated without priority differentiation, then allocation simplicity is maintained, but resource allocation efficiency decreases when resources are limited
Solution Approach 1:
The system segments users into priority-based groups, which creates a structured allocation framework. While this adds some complexity compared to uniform allocation, it dramatically improves resource allocation efficiency by ensuring that limited bandwidth resources are directed to high-priority users first. The segmentation approach provides a clear, manageable method for differential allocation that scales well with the number of users.
Data Source
AI summary
Proposed is a flow control method, including: calculating a first quantity of allocable bandwidth resources corresponding to user sets with different priorities respectively in a current resource allocation cycle according to a quantity of bandwidth resources allocated to the user sets with different priorities in a previous resource allocation cycle; further calculating a second quantity of allocable bandwidth resources corresponding to each user in the user sets with different priorities according to the first quantity; and in response to an access request for data initiated by a user received in the current resource allocation cycle, performing access control for the access request based on the second quantity of the allocable bandwidth resources corresponding to the user. In the above process, bandwidth resources can be dynamically adjusted, the bandwidth resources can be rationally allocated, and precise flow control can be achieved to reduce an affected scope when physical resources are limited.