Passive ATM Network Agent Management via Router Querying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network management methods, such as active polling or pinging, are bandwidth-intensive and unable to manage unknown agents, as they require constant communication and directed polling, which can be costly and inefficient, especially in bandwidth-limited networks.
Innovation Solution
Implementing a passive management system where agents send periodic status messages to a host, which queries routers to detect all communicating agents, compares lists, and generates a list of non-communicating agents, reducing bandwidth usage and enabling management of unknown agents through unsolicited status messages and cryptographic associations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If active polling or pinging is used to monitor agent status, then the host can verify agent status, but bandwidth consumption increases due to constant communication between host and agents
Solution Approach 1:
Instead of the host actively polling agents for status information, the patent inverts the approach by having agents send unsolicited status messages to the host. This reversal eliminates the need for continuous bidirectional communication, significantly reducing bandwidth consumption while maintaining reliable status monitoring.
Solution Approach 2:
Agents transmit status messages periodically or event-driven rather than responding to every poll. This periodic action reduces the frequency of communications compared to continuous polling, thereby lowering bandwidth usage while ensuring the host receives timely status updates.
2Reliability
If active polling is used to manage agents, then the host can detect known agents, but the host cannot manage agents it does not know about since it lacks their addresses
Solution Approach 1:
Agents autonomously introduce themselves by sending unsolicited status messages that include their identity and address information. This self-service approach enables the host to automatically discover and manage previously unknown agents without requiring prior knowledge of their addresses or manual configuration.
Solution Approach 2:
The system implements feedback mechanisms where agents provide status information and the host uses this feedback to dynamically update its agent registry. This feedback loop enables continuous discovery and management of both known and unknown agents, enhancing system adaptability.
3Loss of information
If the host sends polls or pings to agents, then status information can be obtained, but a great deal of bandwidth is required for each poll-response message pair
Solution Approach 1:
The patent extracts the status information transmission from the poll-response cycle. Instead of requiring a complete request-response pair, agents independently transmit status information to the host, eliminating the redundant poll message and reducing bandwidth consumption by approximately 50% compared to traditional polling.
4Reliability
If directed polling is implemented for network management, then known agents can be monitored, but the system becomes complex and costly due to required address knowledge and communication infrastructure
Solution Approach 1:
Agents autonomously provide their own management information by sending unsolicited status messages containing their identity, address, and operational status. This eliminates the need for complex address management systems and reduces infrastructure requirements, simplifying the overall network management architecture.
Data Source
AI summary
Methods of managing a network comprising a plurality of agents. The methods may comprise the step of receiving a periodic status message from each of a first group of agents selected from the plurality of agents The methods may also comprise the steps of querying a router on a local portion of the network positioned to handle communications between components on the local portion of the network and components on a remote portion of the network, and receiving from the router an indication of a second group of agents from the plurality of agents that are communicating through the router. In addition, the methods may comprise the steps of comparing the first group of agents to the second group of agents, and generating a list of agents communicating through the router from which periodic status messages are not being received.