Bandwidth Allocation for Network Clients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network congestion management protocols, such as TCP, lead to unfair bandwidth allocation among users, particularly in large-scale distributed systems, where users with more connections can access disproportionately more bandwidth, creating an imbalance.
Innovation Solution
A system and method that predict demand and allocate bandwidth to each user based on a desired policy, classifying remote clients into individual and group classifications, using weight values and traffic thresholds to distribute bandwidth fairly across the network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If TCP protocol is used for congestion control, then network stability is maintained, but bandwidth allocation fairness deteriorates
Solution Approach 1:
The patent segments bandwidth allocation by classifying remote clients into different groups (first classification and second classification) based on their network traffic estimates. This segmentation allows differential treatment where some clients receive individual bandwidth allocation while others receive group-based allocation, thereby achieving fairness while maintaining overall network stability through structured control mechanisms.
Solution Approach 2:
The patent applies local quality by providing customized bandwidth allocation strategies to different client groups based on their specific characteristics. Clients in the first classification receive individualized bandwidth allocation tailored to their needs, while clients in the second classification receive group-based allocation, allowing the system to optimize fairness for each local group while maintaining global stability.
2Manufacturing precision
If individual bandwidth allocation is provided to all remote clients, then bandwidth allocation fairness is improved, but system complexity increases
Solution Approach 1:
The patent reduces system complexity by segmenting the client population into two classifications. Instead of implementing complex individualized allocation for all clients, the system applies simple group-based allocation to the second classification while reserving individualized allocation only for the first classification, thereby achieving fairness where needed while minimizing overall system complexity.
Solution Approach 2:
The patent applies partial action by providing individualized bandwidth allocation only to clients in the first classification who have specific bandwidth requirements, rather than to all clients. This partial application of the more complex allocation strategy reduces system complexity while still achieving fairness for the clients who need it most.
3Ease of operation
If bandwidth is allocated based on number of connections, then ease of operation is improved, but bandwidth allocation fairness deteriorates
Solution Approach 1:
The patent changes the allocation parameter from simply counting connections to classifying clients based on their network traffic estimates. This parameter change allows the system to maintain operational simplicity through classification-based rules while achieving fairness by allocating bandwidth according to actual traffic needs rather than connection count.
Solution Approach 2:
The patent applies local quality by creating different allocation regimes for different client groups. Clients in the first classification receive individualized allocation based on their specific traffic patterns, while clients in the second classification receive group-based allocation, allowing the system to balance simplicity and fairness locally for each group.
Data Source
AI summary
A system and method are provided that allocate bandwidth on a global large scale network. Bandwidth allocation is based on the predicted bandwidth demands of the network users. Each user may be assigned one of a plurality of different values that determines the amount of bandwidth allocated to that user. In instances where an application runs on behalf of a plurality of remote clients, a system and method is provided that allows for the allocation of bandwidth based each individual remote client.