SDN Controller Subclusters for Distributed Data Center Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a multi-cloud environment, managing network communications across geographically distributed data centers is inefficient and costly due to the need for each data center to have its own Software-Defined Networking (SDN) controller, which consumes limited compute resources and can lead to undesirable routing configurations, impacting performance and reliability.
Innovation Solution
Implementing a subcluster of control nodes within a primary data center to manage and facilitate virtual networks across remote data centers, where the primary data center executes an SDN controller cluster, providing shared resources for analytics and database management, and dedicated control nodes handle route information and communication tunnels, preventing direct peering between subclusters to optimize routing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If each data center has its own SDN controller, then network control is distributed and local, but compute resources are consumed and routing configurations become complex
Solution Approach 1:
The patent segments the SDN controller into a distributed controller architecture where control functions are divided between central controller and local controllers. The central controller manages global routing and policy, while local controllers handle local network operations. This segmentation allows remote data centers to have minimal local control presence while still achieving localized network management, reducing compute resource consumption compared to full controllers at each site.
Solution Approach 2:
The patent introduces intermediary control nodes that act as mediators between the central controller and remote compute nodes. These intermediary nodes receive routing information from the central controller and distribute it to local compute nodes, enabling indirect control. This intermediary layer allows the system to maintain centralized control logic while achieving distributed execution, reducing the need for full SDN controllers at each remote data center.
2Ease of operation
If each data center has its own SDN controller, then local network management is enabled, but undesirable routing configurations occur impacting performance
Solution Approach 1:
The patent implements feedback mechanisms where the central controller receives routing information and status updates from local controllers and compute nodes. Based on this feedback, the central controller dynamically adjusts routing policies and configurations. This feedback loop ensures that routing configurations are optimized based on actual network conditions, preventing undesirable routing decisions while maintaining local management capabilities.
Solution Approach 2:
The patent creates a universal control framework where the central controller serves multiple functions: global routing management, policy enforcement, and coordination with local controllers. This multi-functional approach allows a single centralized controller to manage routing for multiple remote data centers, ensuring consistent and optimal routing configurations across the entire network without requiring separate controllers at each site.
3Adaptability or versatility
If full SDN controller is deployed at remote data center, then complete network control is achieved, but resource consumption increases
Solution Approach 1:
The patent applies partial action by deploying minimal control functionality at remote data centers rather than full SDN controllers. Local controllers perform only essential functions such as receiving routing information from the central controller and forwarding it to compute nodes. This partial deployment provides sufficient network control capability for local operations while significantly reducing compute resource consumption compared to full controllers.
Solution Approach 2:
The patent merges control functions by consolidating routing management and policy enforcement at the central controller, while local controllers handle only basic forwarding and local coordination tasks. This merging of control functions reduces the computational burden at remote data centers while maintaining network control capability. The central controller acts as a shared resource for multiple remote sites, improving overall resource utilization efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure describes examples where a first data center includes a first gateway router, a first set of computing devices, and a second set of computing devices. The first set of computing devices is configured to execute a software defined networking (SDN) controller cluster to facilitate operation of one or more virtual networks within the first data center. The second set of computing devices is configured to execute one or more control nodes to exchange route information, between the first gateway router and a second gateway router of a second data center different than the first data center, for a virtual network between computing devices within the second data center, and to communicate control information for the second data center to the second set of computing devices, wherein the one or more control nodes form a subcluster of the SDN controller cluster.