Scatter Gather Scheme for Aggregating Point-to-Point Interfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network routing and switching devices face bandwidth limitations in point-to-point interfaces, leading to data bottlenecks, especially when handling high-bandwidth applications like streaming video and audio, as existing solutions are difficult to implement or provide limited utility.
Innovation Solution
A load-scattering scheme is employed, where incoming data packets are divided into chunks and distributed across multiple point-to-point interfaces based on the fullness status of FIFOs, using control words to reassemble packets, thereby aggregating bandwidth and balancing load across available channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple point-to-point interfaces are used to increase bandwidth, then data throughput is improved, but interface complexity and synchronization difficulty increase
Solution Approach 1:
The patent segments incoming data packets into multiple data chunks and distributes them across multiple point-to-point interfaces (e.g., interfaces 245-1 through 245-4) using a scatter module. Each interface independently transmits its assigned chunks to corresponding receive interfaces, which then reassemble the original packets using gather modules and control words. This segmentation enables bandwidth aggregation while maintaining manageable interface complexity through standardized chunk handling procedures.
Solution Approach 2:
The patent introduces control words as intermediary elements that facilitate communication between the scatter module and gather modules. Each data chunk is accompanied by control words containing identifiers and sequence information that enable the receive-end gather modules to correctly reassemble packets from multiple interfaces. This intermediary mechanism coordinates the distributed transmission without requiring complex direct synchronization between all interface pairs.
2Productivity
If data is transmitted through multiple channels to avoid bottlenecks, then bandwidth is increased, but load balancing difficulty increases
Solution Approach 1:
The scatter module dynamically selects which point-to-point interface receives each data chunk based on real-time conditions such as interface availability, current load, and FIFO buffer status. This dynamic allocation allows the system to adapt to changing network conditions and balance load across multiple channels automatically, rather than using static assignment. The gather modules correspondingly adapt to receive chunks from variable sources and reassemble packets maintaining correct sequence.
3Speed
If point-to-point interfaces operate at high speed to handle incoming data, then data transfer rate is improved, but data bottleneck occurs at the interface
Solution Approach 1:
The patent merges the bandwidth capacity of multiple point-to-point interfaces by distributing data chunks across all available interfaces simultaneously. Instead of relying on a single high-speed interface that becomes a bottleneck, the system combines the throughput of multiple interfaces (e.g., four 800 Mb/s interfaces providing aggregate bandwidth equivalent to a single 3.2 Gb/s interface). This merging approach eliminates the bottleneck by providing parallel data paths while maintaining high overall transfer rates.
Data Source
AI summary
A mechanism for combining plurality of point-to-point data channels to provide a high-bandwidth data channel having an aggregated bandwidth equivalent to the sum of the bandwidths of the data channels used is provided. A mechanism for scattering segments of incoming data packets, called data chunks, among available point-to-point data channel interfaces is further provided. A decision as to the data channel interface over which to send a data chunk to can be made by examining a fullness status of a FIFO coupled to each interface. An identifier of a data channel on which to expect a subsequent data chunk can be provided in a control word associated with a present chunk of data. Using such information in control words, a receive-end interface can reassemble packets by looking to the control word in a currently processing data chunk to find a subsequent data chunk.