Crossbar Priority Matrix Arbitration for Scalable Fair Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing crossbar circuitry is inefficient in managing adaptive priority arbitration schemes due to the complexity and power consumption associated with large-scale implementations, and it struggles to support scalable and high-speed arbitration mechanisms.
Innovation Solution
The use of a matrix of priority bits within an arbitration circuit to represent and update priority relationships between data source circuits, allowing for efficient and scalable implementations of adaptive priority arbitration schemes, including least recently granted, round robin, and other schemes, by performing operations on selected rows and columns of the matrix.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional crossbar circuitry is used to support adaptive priority arbitration schemes, then arbitration functionality is provided, but device complexity and power consumption increase rapidly with size
Solution Approach 1:
The crossbar circuit is divided into multiple blocks, each with its own local arbitration circuit. Each block independently manages priority arbitration for its subset of input paths, breaking down the complex global arbitration into simpler local decisions. This segmentation reduces overall circuit complexity while maintaining adaptive priority arbitration capabilities across the entire crossbar.
Solution Approach 2:
Priority information is represented using a matrix structure where priority relationships between input paths are encoded in a two-dimensional format. This matrix representation allows efficient storage and manipulation of priority information, enabling adaptive priority schemes without proportionally increasing circuit complexity.
2Adaptability or versatility
If traditional crossbar circuitry is used to support adaptive priority arbitration schemes, then arbitration functionality is provided, but power consumption increases significantly
Solution Approach 1:
By segmenting the crossbar into blocks with local arbitration, power consumption is distributed and reduced. Each local arbitration circuit only needs to manage priority decisions for its specific block rather than the entire crossbar, significantly reducing the power required for priority management while maintaining adaptive arbitration functionality.
3Device complexity
If fixed priority arbitration scheme is used, then arbitration is simple to implement, but high priority data sources may starve lower priority data sources
Solution Approach 1:
The arbitration scheme transitions from static fixed priority to dynamic adaptive priority. Priority values are not fixed but can change over time based on system state and previous processing activity. This allows the system to maintain simple arbitration logic while achieving fair resource distribution through dynamic priority adjustments, preventing starvation of lower priority sources.
Solution Approach 2:
The adaptive priority arbitration incorporates feedback mechanisms where priority decisions are based on previous processing activity and system state. This feedback allows the system to adjust priorities dynamically, ensuring that data sources that have not recently accessed the crossbar can obtain fair access opportunities, preventing starvation while maintaining implementation simplicity.
4Quantity of substance
If crossbar size increases to interconnect large numbers of sources and destinations, then connectivity capacity increases, but complexity grows rapidly making known techniques impractical
Solution Approach 1:
The crossbar is divided into multiple manageable blocks, each handling a subset of connections. This segmentation allows the system to scale to large numbers of sources and destinations by distributing the arbitration logic across multiple local circuits rather than requiring a single complex global arbitration mechanism, making large-scale implementations practical.
Solution Approach 2:
The use of matrix structures to represent priority information provides a scalable organization that grows efficiently with crossbar size. This two-dimensional representation allows the system to handle increasing numbers of inputs and outputs without proportionally increasing arbitration complexity, enabling practical implementation of large-scale crossbars.
Data Source
AI summary
Interconnect circuitry 2 has a plurality of data source circuits 8 connected to respective input paths 4 and a plurality of data destination circuits 10 connected to respective output paths 6. Connection cells 12 provide selective connections between input paths 4 and output paths 6. Arbitration circuitry 26 provides adaptive priority arbitration between overlapping requests received at different input paths. Priority bits 16 within a matrix of priority bit 46 for each output path 10 are used to represent the priority relationships between different input paths which compete for access to that output path 10. Update operations are applied on a per row or per column basis within the matrix to implement update schemes such as least recently granted, most recently granted, round robin, reversal, swap, selective least recently granted, selective most recently granted etc.