Hierarchical Network Traffic Scheduling via Dynamic Node Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hierarchical schedulers in network devices face inefficiencies due to static memory allocation and inability to dynamically compute weights and rate credits for shared intermediate nodes, leading to underutilization of resources and limited scalability.
Innovation Solution
Implementing dynamic node weighting techniques that allow for the dynamic computation of weights and rate credits for shared intermediate nodes, enabling flexible queue assignment and efficient use of on-chip memory by grouping queues into virtual subscribers, thereby reducing the need for static memory allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If static memory allocation is used for hierarchical schedulers, then device complexity is reduced, but adaptability and scalability are limited
Solution Approach 1:
The patent implements dynamic node weighting where intermediate node weights are computed based on the number of virtual subscribers and their individual weights. This dynamic computation allows the scheduler to adapt to changing network conditions and subscriber configurations without requiring static memory allocation for all possible scenarios, thereby improving adaptability while managing complexity through algorithmic approaches.
Solution Approach 2:
The system changes parameters dynamically by computing weights and rate credits based on current subscriber configurations. The weight of an intermediate node is calculated as a function of the number of virtual subscriber child members and their respective weights, allowing the system to adapt to different network topologies and subscriber counts without requiring predetermined static allocations.
2Productivity
If static memory allocation is used for queue configuration, then ease of manufacture is improved, but productivity and resource utilization are reduced
Solution Approach 1:
The patent employs dynamic computation of weights and rate credits that adapts to the actual number and configuration of virtual subscribers. This dynamic approach allows the system to optimize resource allocation and improve productivity by utilizing memory and processing resources based on real-time conditions rather than fixed predetermined configurations.
Solution Approach 2:
The hierarchical scheduler computes its own weight assignments and rate credit allocations based on the current configuration of virtual subscribers and queues. This self-service capability allows the system to automatically optimize its own performance without requiring external intervention or complex manual configuration, thereby improving productivity while simplifying deployment.
3Measurement precision
If shared intermediate nodes are used for multiple virtual subscribers, then device complexity is reduced, but measurement precision and scheduling accuracy are compromised
Solution Approach 1:
The patent calculates precise weight values for shared intermediate nodes based on the number of virtual subscribers and their individual weights. By using mathematical computations to determine exact weight proportions, the system maintains scheduling precision and accuracy even when multiple virtual subscribers share common intermediate nodes, thereby achieving both precision and reduced complexity.
Solution Approach 2:
The system incorporates feedback mechanisms where the weight computations are based on the actual configuration of virtual subscribers and queues. This feedback approach ensures that the scheduling decisions are accurate and reflect the current network state, maintaining precision while the shared node structure reduces overall system complexity.
4Adaptability or versatility
If dynamic node weighting is implemented, then adaptability and scalability are improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent implements dynamic node weighting that computes weights based on the number of virtual subscribers and their individual weights. This dynamic approach enables the system to scale efficiently by adapting to varying numbers of subscribers without requiring proportional increases in system complexity, as the computations are performed algorithmically rather than requiring complex hardware configurations.
Data Source
AI summary
The techniques may provide a hierarchical scheduler for dynamically computing rate credits when a plurality of queues share an intermediate node. For example, the hierarchical scheduler may group respective sets of queues with respective virtual subscribers to be associated with a shared intermediate node. The weight used by the shared intermediate node may be computed as a function of the number of virtual subscriber child members of the shared intermediate node and their respective weights to correctly proportion the services to the queues. The techniques may also provide a hierarchical scheduler for dynamically computing rate credits allocated to queues associated with a shared intermediate node. For example, the number of rate credits allocated to a queue for a virtual subscriber is based on the product of the virtual subscriber weight and a queue weighted fraction of the queues for the virtual subscriber.