NoC Bandwidth Arbiter Dynamic Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bandwidth allocation methods in System-on-Chip (SoC) technologies face challenges due to non-ideal behavior of real systems, such as non-uniform traffic, interference, and variations in storage efficiency, leading to inefficiencies and inability to meet the requirements of all initiator modules effectively.
Innovation Solution
A robust bandwidth recovery mechanism that dynamically adapts allocation parameters to address unpredictability, ensuring that bandwidth is allocated efficiently by adjusting priorities and transaction frequencies, allowing for complete utilization of available bandwidth while minimizing deviations from requested values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a round-robin end-to-end arbitration scheme is implemented with fixed bandwidth requests at input points, then bandwidth allocation can be controlled at the source, but the system cannot effectively handle non-uniform traffic, interference, and variations in storage efficiency
Solution Approach 1:
The patent implements a feedback mechanism where the arbiter continuously monitors actual bandwidth usage and adjusts arbitration decisions in real-time. The system measures the actual bandwidth received by each initiator and compares it with the requested bandwidth, then dynamically adjusts the arbitration policy to compensate for deviations caused by non-uniform traffic, interference, and storage efficiency variations. This closed-loop control ensures reliable bandwidth allocation despite real-world imperfections.
Solution Approach 2:
The arbitration scheme transitions from a static round-robin approach to a dynamic adaptive arbitration mechanism. The system continuously adjusts arbitration weights, priorities, and timing based on real-time measurements of actual bandwidth usage. This dynamic adaptation allows the system to respond to changing traffic patterns, interference conditions, and storage efficiency variations, maintaining allocation accuracy under varying operational conditions.
2Productivity
If priority adjustment is used to control bandwidth allocation, then bandwidth can be increased for high-priority initiators, but this creates interference and non-uniform traffic patterns that reduce overall system efficiency
Solution Approach 1:
The patent dynamically changes multiple arbitration parameters including priority levels, arbitration weights, transaction timing, and inter-transaction intervals. Instead of relying solely on priority adjustment, the system modifies multiple parameters in combination to achieve bandwidth allocation while maintaining traffic uniformity. This multi-parameter control allows the system to satisfy priority requirements without creating harmful interference patterns.
Solution Approach 2:
The system implements periodic measurement and adjustment cycles where bandwidth usage is monitored over defined intervals, and arbitration parameters are updated periodically based on accumulated measurements. This periodic action smooths out traffic patterns and prevents the formation of interfering non-uniform traffic, as adjustments are made systematically rather than reactively to instantaneous conditions.
3Device complexity
If the arbitration mechanism operates with fixed parameters, then system design is simplified, but it cannot adapt to unpredictable behavior of applications and real-world non-ideal conditions
Solution Approach 1:
The arbitration mechanism implements self-service through automatic measurement, monitoring, and adjustment of arbitration parameters based on real-time system conditions. The system autonomously detects bandwidth usage patterns, identifies deviations from allocated bandwidth, and adjusts arbitration decisions without external intervention. This self-adjusting capability provides adaptability to unpredictable application behavior while maintaining relatively simple system architecture, as the complexity is confined to the autonomous control logic rather than the overall system design.
Data Source
AI summary
A system comprises a resource, such as an interconnection, for example, of the Network-on-Chip (NoC) type, having an overall bandwidth available for allocation to a set of initiators that compete for allocation of the overall bandwidth. The system includes a communication arbiter for allocating the overall bandwidth to the initiators according to respective values of bandwidth requested (RBW) by the initiators. A control device (50) is configured to detect the deviation between the value of bandwidth allocated to the initiators and the respective value of requested bandwidth and allocate the overall bandwidth to the initiators in a dynamic way minimizing the mean value of the deviation.


