Distributed Management for Virtual Network Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Centralized management and orchestration of virtual network functions in communication networks is vulnerable to single points of failure and does not support network partitioning, leading to potential network disruptions due to failures or attacks.
Innovation Solution
A distributed management system is implemented across multiple points of presence (PoPs) within the network, where each PoP includes a management and orchestration module that manages and orchestrates virtual network functions, using a distributed routing table and algorithms for instance placement and chaining to ensure resilient network operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a centralized MANO server is used to manage virtual network functions, then management and orchestration can be centralized and simplified, but the system becomes vulnerable to single points of failure and cannot support network partitioning
Solution Approach 1:
The centralized MANO server is segmented into multiple distributed MANO servers deployed across different network domains. Each MANO server manages virtual network functions within its local domain, eliminating the single point of failure while maintaining manageable operations through modular domain-based control
2Device complexity
If a centralized MANO server is used, then orchestration decisions can be made centrally, but the network cannot be partitioned into subnets without losing connectivity to the MANO server
Solution Approach 1:
The network is segmented into multiple autonomous domains, each with its own MANO server capable of independent orchestration decisions. This allows the network to be partitioned into subnets while maintaining full orchestration functionality within each domain, eliminating dependency on a central server for network resilience
Solution Approach 2:
Each domain is equipped with local MANO server capabilities that provide orchestration functions tailored to local needs. This local autonomy enables domains to operate independently during network partitions while maintaining appropriate orchestration control for local virtual network functions
3Reliability
If management functions are distributed across multiple PoPs, then network resilience is improved and partitioning is supported, but system complexity increases
Solution Approach 1:
The distributed MANO servers are organized into hierarchical domains with clear boundaries and responsibilities. Each domain server manages a specific set of virtual network functions and resources, reducing the complexity burden on individual servers while maintaining overall system resilience through the segmented architecture
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This system includes a virtual network function executed by a point of presence (5, 6, 7) comprising: a routing module (40) connected between a physical port (26) and an internal port (41), the routing module exchanging routing information with other machines on the network; a switching module (50) between the internal port (41) and a plurality of virtual input ports (61, 62, 63); a virtualization module (60); and an instance (71, 72, 73) of the virtual network function, the execution of which is managed by the virtualization module (60), a single virtual input port being associated with said instance, the switching module directing a packet from the internal port to the virtual input port of the instance of the virtual network function for processing said packet, the software layer (30) further comprising a management and orchestration module (80).