Arbitration of Data Transfer Requests Using Priority Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transfer request arbitration methods in mesh-based networks fail to efficiently manage traffic congestion and packet forwarding, leading to potential buffer overflows and unfair selection among nodes, especially when multiple nodes are equally congested.
Innovation Solution
A credit value-aware, traffic congestion-aware, and fair packet arbitration mechanism that uses priority matrices to select the most congested node while ensuring equal priority relations between nodes, implemented in an arbiter with a credit gate, traffic matrices, and priority matrices to manage packet forwarding and acknowledge requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If priority-based arbitration is used to select nodes, then packet forwarding speed is improved, but fairness among equally congested nodes deteriorates
Solution Approach 1:
The priority matrix is periodically inverted to change the priority relationships between nodes. This periodic inversion ensures that over time, all nodes get equal opportunity to be selected, achieving fairness while maintaining fast packet forwarding through priority-based selection at each moment.
Solution Approach 2:
The arbitration mechanism dynamically adjusts priority assignments by inverting the priority matrix based on node credit values. This dynamic adjustment allows the system to adapt to changing network conditions while maintaining both speed and fairness through time-varying priority relationships.
2Productivity
If nodes with higher priority are selected for packet forwarding, then throughput is improved, but buffer overflow risk increases due to unfair selection
Solution Approach 1:
The arbiter uses feedback from credit values held by nodes to dynamically adjust priority assignments. Nodes with lower credit values receive higher priorities, creating a feedback mechanism that prevents buffer overflow while maintaining high throughput through efficient packet forwarding.
Solution Approach 2:
Periodic inversion of the priority matrix ensures that no single node maintains high priority indefinitely. This periodic change prevents sustained unfair selection that could lead to buffer overflow, while maintaining high throughput through priority-based arbitration at each time step.
3Reliability
If credit value-aware arbitration is implemented, then buffer overflow prevention is improved, but arbitration complexity increases
Solution Approach 1:
Nodes perform preliminary actions by maintaining credit values that indicate their buffer status before arbitration occurs. The arbiter uses these pre-computed credit values to make fair priority assignments, preventing buffer overflow without requiring complex real-time analysis during the arbitration process.
Solution Approach 2:
The system changes the parameter of priority assignment based on credit values. By inverting priorities according to credit value thresholds, the system achieves buffer overflow prevention through a relatively simple parameter transformation rather than complex arbitration logic.
Data Source
AI summary
A method for arbitrating data transfer requests from a plurality of nodes includes specifying one or more nodes among the plurality of nodes, the one or more nodes satisfying a predetermined condition, and selecting, if two or more nodes are specified among the plurality of nodes, one node from the two or more nodes using priority information, the priority information indicating correspondence between the plurality of nodes and a plurality of priorities each assigned to one of the plurality of nodes, the correspondence changing so that the plurality of priorities are assigned equally to each of the plurality of nodes and high and low relations appear equally between pairs of priorities each assigned to a pair of nodes of the plurality of nodes.