SD-WAN Controller Load Balancing via Dynamic OMP Shedding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
SD-WANs face challenges in maintaining balance and stability due to control plane disruptions, leading to overloaded and unstable controllers, which existing technologies fail to address automatically and effectively.
Innovation Solution
The implementation of a system and method that dynamically processes load information digest messages to perform Overlay Management Protocol (OMP) load shedding, allowing for the selection and assignment of centralized controllers to edge devices based on load requirements and controller group preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If edge devices use static hashing to select controllers, then deterministic connection behavior is achieved, but controller load distribution and peer distribution are not optimized
Solution Approach 1:
The patent transitions from static hashing to dynamic controller selection. The orchestrator continuously monitors controller load and peer distribution, then dynamically assigns edge devices to controllers based on current system state. This dynamic approach allows the system to adapt to changing conditions while maintaining stable connections through orchestrated reassignments.
Solution Approach 2:
The system implements feedback loops where the orchestrator continuously collects information about controller load, peer distribution, and edge device connections. Based on this feedback, the orchestrator makes informed decisions about controller assignments and performs load balancing operations to maintain optimal system state.
2Reliability
If edge devices reconnect to operational controllers after control plane disruption, then connection continuity is restored, but operational controllers become overloaded and unstable
Solution Approach 1:
The orchestrator performs preliminary actions by proactively detecting control plane disruptions and managing reconnection processes before controllers become overloaded. When a disruption is detected, the orchestrator coordinates reconnections and performs load shedding in advance to prevent instability, rather than reacting after overload occurs.
Solution Approach 2:
The orchestrator acts as an intermediary between edge devices and controllers during reconnection events. It receives reconnection requests from edge devices, evaluates current controller states, and directs devices to appropriate controllers. This intermediary role prevents direct uncoordinated reconnections that would cause overload.
3Ease of operation
If centralized controllers handle all control plane functions, then centralized management is achieved, but system vulnerability to overload and failure increases
Solution Approach 1:
The patent segments the control plane functions across multiple centralized controllers rather than concentrating all functions in a single controller. Each controller handles a portion of the control plane workload, and the orchestrator manages distribution among them. This segmentation provides redundancy and prevents single-point failures while maintaining centralized management through the orchestrator.
Data Source
AI summary
Systems and methods are provided for re-balancing and healing of an SD-WAN in an unbalanced state and/or experiencing one or more failure states. In response to a request to connect to a new controller resulting from OMP load shedding from a first controller, the system can identify other controllers capable of handling the load requirements of the edge router. The system can incorporate the controller group preference of the edge router and select a second controller based on the identified other controllers and within the preferred controller group. If not possible, the system can temporarily assign the edge router to non-preferred controller groups and move them back to controllers in the preferred controller group once it becomes viable. The system further enhances OMP graceful restart (GR) logic to incorporate the load shedding effect and avoid unnecessary route retention that GR entails.