Traffic Assessment in Multiple Stage Switch Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multiple stage switch networks, such as Clos networks, face resource and capacity constraints that hinder efficient traffic routing, leading to potential network congestion and unnecessary investment in new hardware due to improper load balancing.
Innovation Solution
An apparatus and method for assessing traffic through multiple stage switch networks by calculating network resource requirements based on specific traffic types, using an algorithm that generalizes to any route, allowing for comparison with network capacity and modification of switches or traffic flow to optimize resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple stage switch networks are used to increase traffic moving capability, then the ability to move traffic is improved, but resource or capacity constraints prevent optimal utilization
Solution Approach 1:
The patent applies preliminary action by calculating network resource requirements before routing traffic through the multiple stage switch network. The system determines the resource requirements of traffic based on its type and compares them against available capacity in advance, ensuring that routing decisions are made only when capacity constraints are satisfied. This prevents overloading the network while maximizing traffic movement capability.
2Productivity
If load balancing is not properly implemented between switches, then network congestion occurs and hardware sits underutilized, but implementing load balancing requires additional configuration complexity
Solution Approach 1:
The patent implements self-service by enabling the network system to automatically calculate and distribute traffic based on resource requirements and available capacity. The system autonomously determines optimal routing paths without requiring manual load balancing configuration, thereby improving hardware utilization while avoiding the complexity of manual load balancing setup.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting routing decisions based on calculated resource requirement parameters and capacity parameters. The system changes routing parameters in real-time to balance load across switches, improving utilization efficiency without requiring complex static configuration.
3Reliability
If network capacity is not assessed before routing traffic, then traffic may exceed network capacity causing congestion, but assessing capacity requires additional processing time
Solution Approach 1:
The patent applies preliminary action by performing capacity assessment and resource requirement calculation before traffic routing decisions are made. This ensures that only traffic within capacity constraints is routed, maintaining reliability while the efficient calculation method minimizes the time penalty.
4Productivity
If existing network hardware is not optimally configured, then productivity improvements cannot be achieved, but investment in new hardware is unnecessary
Solution Approach 1:
The patent implements self-service by enabling the network system to automatically optimize its own configuration through resource requirement calculation and capacity-based routing decisions. This achieves productivity improvements through intelligent software-based optimization rather than requiring complex manual reconfiguration or hardware upgrades.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An apparatus is provided for implementation of a system for assessing traffic through a multiple stage switch network. The apparatus may receive an indication of traffic to be routed through a multiple stage switch network configured to route the traffic from one or more input ports to one or more output ports according to a type of the traffic. The apparatus may calculate a network resource requirement of the traffic based on the input port(s), output port(s) and type of the traffic. Here, the network resource requirement may be calculated according to an algorithm specific to the type of the traffic, but generalized to any specific route of the traffic through the multiple stage switch network. And the apparatus may assess the traffic based on a comparison of a capacity of the multiple stage switch network, and the network resource requirement of the traffic.