SDN Sampling Node Traffic Flow Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Software Defined Networks (SDNs), managing dynamic and varying traffic patterns for value-added services and Network Functions Virtualizations (NFVs) is costly and prone to overload, as existing solutions fail to efficiently route traffic flows based on capacity demands.
Innovation Solution
A sampling node in an SDN that receives a fraction of total traffic flows, identifies which flows benefit from value-added services, determines individual flow capacity demands, and selectively routes them through service optimizing nodes based on available capacity, informing the flow switch which flows to bypass or route via these nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the service optimising node is designed to handle dynamically and fast changing traffic patterns, then the node can adapt to varying capacity demands, but the cost of designing and provisioning such a node becomes excessive
Solution Approach 1:
The patent applies preliminary action by pre-provisioning the service optimising node with a fixed, maximum capacity configuration. Instead of dynamically adapting the node's capacity, the system proactively allocates sufficient resources in advance to handle peak traffic patterns, eliminating the need for complex real-time capacity adjustment mechanisms
Solution Approach 2:
The sampling node acts as an intermediary that analyzes traffic patterns and selectively routes only those flows exceeding a threshold to the service optimising node. This intermediary function protects the service optimising node from being overwhelmed by all incoming traffic, allowing it to be provisioned with lower, more cost-effective capacity while still handling peak demands when needed
2Productivity
If all traffic flows are routed via the service optimising node, then value added services can be provided to all flows, but the node becomes overloaded and performance degrades
Solution Approach 1:
The patent applies local quality by selectively routing only specific traffic flows that benefit from value added services and exceed a capacity threshold through the service optimising node. Other flows are routed directly without optimization, distributing the routing decision locally at each flow level rather than applying a uniform routing policy to all traffic
Solution Approach 2:
The system implements partial action by routing only a subset of traffic flows through the service optimising node rather than all flows. The sampling node identifies and selects only those flows that meet specific criteria (capacity threshold and service benefit), applying optimization partially rather than universally, thus preventing node overload while still providing value added services where beneficial
3Reliability
If the service optimising node is designed with high capacity to handle peak traffic, then overload situations are avoided, but capacity is wasted during low traffic periods
Solution Approach 1:
The sampling node implements partial action by routing only the necessary portion of traffic flows through the service optimising node - specifically those flows that exceed the capacity threshold and would benefit from optimization. This selective routing ensures the node processes only what is needed, avoiding the waste of handling unnecessary traffic during low-demand periods while maintaining reliability for critical flows
4Productivity
If traffic routing is dynamically adjusted based on capacity demand, then capacity usage is maximized, but complex supervision and monitoring are required
Solution Approach 1:
The sampling node implements self-service by autonomously analyzing traffic flow characteristics, determining capacity demands, and making routing decisions without requiring external supervision or monitoring. The node independently evaluates each flow against predefined criteria and automatically routes appropriate flows to the service optimising node, eliminating the need for complex external control mechanisms
Solution Approach 2:
The system implements feedback through the sampling node's continuous monitoring of traffic flow capacity demands and its automatic adjustment of routing decisions based on this information. The node uses the capacity demand data as feedback to dynamically control which flows are routed through the service optimising node, maximizing capacity usage while maintaining simple operation through automated closed-loop control
Data Source
AI summary
A sampling node in a SDN and a method performed thereby for handling flows through the SDN between client(s) and origin server(s) of a communication network connected to the SDN are provided. The method comprising receiving (110) a fraction of a total amount of traffic flows originating at client(s) served by the SDN, and destined for the origin server(s); identifying (120) which of the received traffic flows that benefit from being routed via a service optimising node, capable of providing value added services, VAS, to the traffic flows, by fulfilling predetermined conditions; and determining (130), for each individual traffic flow, a capacity demand of the flow. The method further comprises selecting (150) which traffic flows that shall be routed via the service optimising node on the basis of their capacity demand considering a capacity of the service optimising node; and informing (160) a flow switch of the SDN about which traffic flows should bypass the service optimising node and which traffic flows that should be routed via the service optimising node.