Guided Network Management via Shadow Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current orchestration systems in service providers, such as AT&T, face frustration due to over-complicated network status readouts, non-specific network manipulations, and the inability to quickly revert changes, leading to inefficiencies in managing complex virtualized networks.
Innovation Solution
A guided network management system that monitors traffic, captures events, creates snapshots, and builds shadow networks to emulate network states, allowing for A/B testing and presenting modification options to mitigate operational issues through a graphical user interface, enabling informed operator decisions and automated evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If orchestration systems perform automated network manipulations, then operational tasks are reduced, but operators cannot quickly revert changes and lack specific control
Solution Approach 1:
The system creates snapshots of network states before performing automated manipulations. This preliminary action captures the current configuration, allowing operators to revert changes by restoring from these pre-manipulation snapshots if needed, thus resolving the contradiction between automation and ease of reversal.
Solution Approach 2:
The system creates shadow networks that are copies of the actual virtualized network. These shadow networks allow operators to test and evaluate modifications in a safe environment before applying them to the production network, enabling easy reversal by simply not deploying changes from the shadow network to the actual network.
2Loss of information
If orchestration systems provide detailed network status readouts, then monitoring capability is improved, but system complexity increases
Solution Approach 1:
The system creates shadow networks that replicate the actual network state. Operators can monitor and analyze network status in the simplified shadow network environment without the complexity of the full orchestration system, thus improving visibility while managing complexity through abstraction.
Solution Approach 2:
The system separates monitoring functions into a distinct shadow network layer, allowing detailed network status monitoring to be performed independently from the complex orchestration logic. This segmentation enables improved visibility without proportionally increasing the perceived system complexity for operators.
3Ease of manufacture
If virtualized networks use COTS hardware platforms, then cost is reduced, but network management complexity increases
Solution Approach 1:
The system creates shadow networks that replicate virtualized network states on COTS hardware. This copying approach allows operators to manage and test network configurations in a simplified shadow environment before deploying to the actual virtualized network, reducing the management complexity inherent in COTS-based virtualized networks.
Solution Approach 2:
The shadow network acts as an intermediary layer between operators and the complex virtualized network running on COTS hardware. This intermediary provides a simplified interface for management and testing, reducing the complexity burden while maintaining the cost benefits of COTS hardware platforms.
Data Source
AI summary
The concepts and technologies disclosed herein are directed, in part, to a system that can monitor traffic traversing a virtualized network that includes a plurality of virtual network functions (“VNFs”) that provide, at least in part, a service. The system can capture an event from the traffic. The event can involve at least one VNF, and can negatively affect at least one operational aspect of the virtualized network in providing the service. The system can create snapshot that represents a network state of the virtualized network during the event. The system can create, based upon the snapshot, a shadow network. The shadow network can include a network emulation of the network state of the virtualized network during the event. The system can determine, from the shadow network, at least one modification to at least a portion of the virtualized network that would at least mitigate negative effects of the event.