Switching Device Segmentation and Aggregation for Bandwidth Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Store-and-forward devices in packet networks face inefficiencies due to variable-sized packets, leading to wasted bandwidth and challenges in separating the crossbar switching matrix and fabric scheduler, especially in high-speed environments where reconfiguration times become a bottleneck.
Innovation Solution
The implementation of a data aggregation scheme that segments variable-size packets into smaller units (segments) and aggregates them into frames for switching, allowing for efficient bandwidth utilization and independent scheduling of the crossbar switching matrix and fabric scheduler.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If packets are transferred in batches through the crossbar switching matrix, then the switching paths are kept unchanged for the duration of the longest packet, but this results in wasted bandwidth when packets are of variable size
Solution Approach 1:
The patent segments variable-size packets into fixed-size cells (e.g., 64-byte cells) for transmission through the crossbar switching matrix. Each packet is divided into multiple cells of uniform size, which are then transmitted in batches. This segmentation allows the crossbar to maintain stable switching paths while eliminating bandwidth wastage, as each cell occupies the full allocated bandwidth regardless of the original packet size.
2Productivity
If the fabric scheduler computes a new schedule during the transmission time of a cell, then scheduling can keep up with high-speed data rates, but this requires extremely fast reconfiguration times that many optical switching devices cannot achieve
Solution Approach 1:
The patent implements preliminary action by having the fabric scheduler compute the entire next schedule in advance, during the transmission of the current batch of cells. The scheduler calculates all switching path configurations for future batches before they are needed, allowing optical switching devices with longer reconfiguration times to catch up. This advance computation decouples the scheduling rate from the reconfiguration speed, enabling high-productivity scheduling even with slower optical crossbars.
3Adaptability or versatility
If the fabric scheduler is physically separated from the crossbar matrix, then independent scheduling is possible, but communication delays between the two become a bottleneck
Solution Approach 1:
The patent introduces an intermediary mechanism where the fabric scheduler communicates switching path configurations to the crossbar matrix in advance, during the transmission of current batches. This intermediary scheduling approach allows the fabric scheduler to be physically separated from the crossbar matrix while minimizing communication delays, as the scheduler prepares configurations ahead of time rather than requiring real-time interaction during cell transmission.
Data Source
AI summary
In general, in one aspect, a switching device is described that includes a segmentation unit to receive packets and divide packets having a length greater than a maximum segment length into multiple segments. A plurality of queues associated with a source and a destination stores the segments. A request generator generates requests that include external factors including amount of data contained in the queue and at least some subset of priority and age. A scheduler receives the requests and assigns the requests an internal priority based on the external factors. The scheduler processes the requests for the queues by internal priority in order to generate grants. A framer, responsive to the scheduler, aggregates a plurality of segments for the queues that received a grant to form a frame and to transmit the frame to an associated destination. The frame may contain segments associated with different packets.