Ultra-high radix communication network switch architecture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scaling up communication networks with multiple layers of switches increases latency, power consumption, physical footprint, and cost, which is problematic for certain applications.
Innovation Solution
A communication network architecture comprising a plurality of first and second switches, where each first switch has integrated circuit (IC) switch chips with network interfaces, downlink, and uplink ports, and each second switch has ports coupled to uplink ports of first switches, IC switch chips, and serializers/deserializers (SERDES) to communicate with external network interfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple layers of switches are added to scale up communication networks, then the number of ports and switching capacity increases, but latency increases
Solution Approach 1:
The network is divided into two layers: first switches with fewer ports handling local connectivity, and second switches with high port counts handling aggregation and backbone traffic. This segmentation allows traffic to be switched locally when possible, reducing the number of hops and latency while still providing scalable capacity through the second layer.
Solution Approach 2:
The patent transitions from a single-layer network architecture to a two-layer hierarchical architecture. This dimensional change in network topology enables simultaneous optimization of local switching speed (first layer) and overall network capacity (second layer), resolving the contradiction between scaling capacity and maintaining low latency.
2Productivity
If multiple layers of switches are added to scale up communication networks, then the number of ports and switching capacity increases, but power consumption increases
Solution Approach 1:
By segmenting the network into two functional layers with different port count characteristics, the system can optimize power consumption for each layer's specific workload. First switches handle lower-volume local traffic with fewer active ports, while second switches handle aggregation traffic, reducing total power consumption compared to a single-layer design where all switches would need high port counts.
Solution Approach 2:
Each layer is designed with quality characteristics suited to its function: first switches have fewer ports optimized for local connectivity, while second switches have high port counts optimized for aggregation. This local quality optimization ensures that power consumption is matched to actual traffic requirements at each layer, avoiding the waste of having all switches uniformly high-capacity.
3Productivity
If multiple layers of switches are added to scale up communication networks, then the number of ports and switching capacity increases, but physical footprint increases
Solution Approach 1:
The network infrastructure is segmented into two tiers with different physical footprint requirements. First switches with fewer ports require less physical space each, while second switches with high port counts consolidate aggregation functions into fewer units. This segmentation reduces the total physical footprint compared to deploying many high-port-count switches throughout the network.
Solution Approach 2:
The first layer switches are nested within the coverage area of second layer switches, creating a hierarchical physical deployment. This nesting allows the network to achieve high switching capacity through functional hierarchy rather than through proportional increases in total physical equipment, reducing overall footprint.
4Productivity
If multiple layers of switches are added to scale up communication networks, then the number of ports and switching capacity increases, but cost increases
Solution Approach 1:
The network is segmented into two layers with different cost characteristics. First switches with fewer ports are less expensive individual units, and second switches with high port counts serve as consolidation points. This segmentation reduces total cost compared to a single-layer design where every switch would need high port counts to provide equivalent capacity, as not all network positions require high-port-count devices.
Solution Approach 2:
Each layer is assigned quality characteristics appropriate to its function: first switches provide basic switching capability at lower cost for local connectivity, while second switches provide high-capacity aggregation at higher cost only where needed. This local quality differentiation optimizes the cost-capacity ratio across the entire network, avoiding the expense of deploying high-end switches everywhere.
Data Source
AI summary
A communication network includes first switches interconnected with second switches. Each first switch includes a first integrated circuit (IC) switch chip, downlink ports, and uplink ports. Each second switch includes ports coupled to at least one uplink port of each of the first switches, and a second IC switch chip in an IC package. To permit each second IC switch chip to forward packets amongst a large number of first switches and to reduce a number of external interconnects of the IC package, each second IC switch chip includes sets of multiplexer/demultiplexer circuitry, each multiplexer/demultiplexer circuitry being coupled between an external interconnect, and a set of multiple internal network interfaces of the second IC switch chip. The multiplexer/demultiplexer circuitry demultiplexes a data stream from the external interconnect to multiple internal network interfaces, and multiplexes multiple data streams from the multiple internal network interfaces to the external interconnect.