Automated Microflow Devolvement for Network Path Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Centralized control plane networking architectures face challenges in maintaining flow granularity for path performance metrics when using aggregate flow entries, as existing solutions either reduce processor involvement or increase network traffic, and lack mechanisms for automatic flow performance metric provision upon termination.
Innovation Solution
The automated devolvement of microflows from aggregate flows, where an ingress node generates microflow entries with idle timeouts and sends path measurement metrics to the controller node upon timeout, allowing for granular performance metric collection without excessive processor involvement or network traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If aggregate flow entries are used to group multiple flows into a single flow entry, then the number of flow entries and controller node processing requirements are reduced, but flow granularity and path performance metric measurement capability are lost
Solution Approach 1:
The patent segments aggregate flows into individual microflows by extracting specific flow identifiers (such as IP addresses, MAC addresses, or port numbers) from packets matching the aggregate flow entry. This segmentation allows the system to maintain granular measurement capability while using aggregate flow entries for routing decisions, resolving the contradiction between reducing flow entry complexity and maintaining measurement precision.
Solution Approach 2:
The patent introduces an intermediary mechanism where datapath nodes generate and track microflow identifiers alongside aggregate flow entries. This intermediary layer enables the system to associate granular flow information with aggregate routing decisions, allowing performance metrics to be collected at microflow level while using aggregate entries for simplified controller processing.
2Measurement precision
If active measurement mechanisms inject probe packets between datapath nodes, then path performance metrics can be measured, but network traffic increases and datapath nodes require additional active measurement functionality
Solution Approach 1:
The patent enables datapath nodes to perform passive self-measurement by utilizing existing packet flow information already present in the network. Nodes automatically generate microflow entries and track performance metrics based on packets they naturally forward, eliminating the need for active probe injection and reducing the complexity of active measurement functionality while maintaining measurement capability.
Solution Approach 2:
The patent discards the need for separate active measurement probe packets by recovering useful measurement information from the actual data packets flowing through the network. By extracting flow identifiers and performance data from normal network traffic, the system achieves measurement functionality without adding separate measurement traffic or complex active measurement hardware.
3Measurement precision
If flows are devolved individually to maintain granularity, then path performance metrics become measurable, but processor involvement and controller node processing requirements increase
Solution Approach 1:
The patent performs preliminary actions by pre-configuring aggregate flow entries with embedded microflow selection criteria and identifiers at datapath nodes. This preliminary setup allows flows to be automatically devolved and tracked without requiring real-time controller intervention for each individual flow, maintaining granularity while reducing controller processing burden through advance preparation of measurement infrastructure.
Solution Approach 2:
The patent applies partial devolvement by selectively creating microflow entries only for flows that require granular measurement, rather than devolving all flows individually. This partial action approach maintains measurement precision for critical flows while avoiding the excessive processing overhead of individual flow devolvement for all traffic, optimizing the balance between granularity and processing efficiency.
Data Source
AI summary
Mechanisms for devolving microflows from aggregate flows are disclosed. An ingress node receives a packet that matches an aggregate flow entry in a flow table. A determination that a devolve action is associated with the aggregate flow entry is made. Based on the determination that the devolve action is associated with the aggregate flow entry, a microflow flow entry is generated in the flow table to define a microflow. The microflow flow entry includes header information extracted from the packet. Microflow generation information that identifies the microflow is sent to a controller node. It is determined that the microflow has timed out based on an idle timeout period of time. In response to determining that the microflow has timed out, microflow termination information that includes path measurement metric information associated with the microflow is sent to the controller node.