A distributed network makes network services available to end users at various nodes or connection points throughout the distributed network’s geographic area. A network administrator monitors the performance, capability, and availability of the distributed network to provide the network services. However, the network administrator may be limited to network traffic or other network-side parameters that may not provide an accurate or a conclusive representation of the state of the distributed network. For example, diminished or decreased network traffic could indicate a malfunction in the distributed network or be a natural consequence of a decreased number of end users. Cost, infrastructure requirements, and other limitations prevent installation and operation of a secondary network, which could be used to conclusively determine the conditions of the area within the distributed network. Instead, machine-learning algorithms may monitor and model some features of the distributed network, which may supplement service availability composite metrics, and allow the network administrator to better evaluate the condition of the distributed network without the need of the secondary network.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Bashyam, Vijayaraghavan and Valliappan, Nachiappan, "Method To Estimate Network Availability", Technical Disclosure Commons, (October 30, 2018)