Distributed Network Management System with Automatic NMS Role Reassignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network management systems are vulnerable to single-point failures and inefficient communication models, particularly in distributed NMS architectures, which can lead to network outages and resource bottlenecks during failures.
Innovation Solution
A distributed network management system where each hardware server includes both NMS and EMS functionality, allowing automatic assignment of NMS roles among servers, and using a star/broadcast method or circular message queue for data sharing, minimizing reliance on a single master server and enhancing fault tolerance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a hierarchical NMS architecture with a central NMS server is used, then centralized network management and alarm status sharing are achieved, but the system creates a single point of failure and processing bottleneck
Solution Approach 1:
The patent segments the centralized NMS functionality by distributing NMS server capabilities to multiple EMS servers throughout the network. Each EMS server can independently provide NMS functions, eliminating the single central NMS server and its associated single point of failure. This segmentation allows the system to maintain centralized management capabilities while improving reliability through distribution.
Solution Approach 2:
The patent transitions from a single-layer hierarchical architecture to a multi-dimensional distributed architecture where NMS functionality exists at multiple levels and locations. EMS servers can operate as NMS servers, creating a multi-dimensional management structure that eliminates the vertical bottleneck of a single central NMS server while maintaining comprehensive network management capabilities.
2Loss of information
If a master server architecture is used in distributed NMS, then alarm status sharing is achieved, but the system creates a single point of failure and communication bottleneck
Solution Approach 1:
The patent merges NMS functionality directly into EMS servers, eliminating the need for a separate master server. Each EMS server with NMS capability can independently share alarm status information with other servers. This merging distributes the information sharing function across multiple nodes rather than concentrating it in a single master server, thereby eliminating the communication bottleneck and single point of failure.
3Adaptability or versatility
If TCP/IP client/server communication model is used, then distributed NMS functionality is achieved, but the system incurs high processing and transmission overhead
Solution Approach 1:
The patent implements multi-functionality by enabling EMS servers to simultaneously perform both EMS and NMS functions. This universal capability allows servers to exchange alarm status information directly without requiring complex client/server communication protocols. By making servers universally capable, the system achieves distributed NMS functionality while reducing communication overhead through direct peer-to-peer interaction.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for method of managing a network. Each of a plurality of servers includes an element management system (EMS) function for managing network elements within the network and a network management system (NMS) function for managing ones of the plurality of servers running the EMS function. An NMS function is activated on one of the servers when the plurality of servers shares a one network domain. When the network domain is divided into multiple domains, e.g. by a network failure, an NMS function is activated on at least one other one of the plurality of servers whereby each of the multiple network domains has at least one associated active NMS function.