Distributed type system network resource operation management reliability raising method

Abstract

The invention discloses a distributed type system network resource operation management reliability raising method, which comprises the following steps: mapping the received tasks according to the use conditions of the resources and the learning experience of a deep learning neural network; based on the cost computation of the deep learning neural network, adjusting the number of the task weights in the mapping; and executing above operations to all tasks and updating the output weight of the deep learning neural network. With the method, under the condition of a hard-to-predict complex scenario, it is possible to not add complexity to the computation, but to adapt for best solutions to the problematic space.

Description

Technical field [0001] The present invention relates to network management technology, in particular to a method for improving the reliability of operation and maintenance management of distributed system network resources. Background technique [0002] Traditional distributed system network resource operation and maintenance monitoring and management focuses on the management of network performance and network failures. The management of network performance is mainly aimed at the analysis of historical data, while the management of network faults is mainly aimed at the real-time monitoring of network elements. The network management system can monitor network elements and network connections, collect various performance data, issue early warnings for failures, and generate data reports for managers' reference. However, the network is only the basis of the application, and simply treating the network as the monitoring object does not provide much intentional and intuitive data. ...

AI Technical Summary

Problems solved by technology

[0004] The main shortcomings of the traditional distributed operation and maintenance monitoring system include: the management structure of the monitoring object is single, there is no division of multi-level directories, and each monitoring and business monitoring are independent of each other and have no relationship

A single list structure will increase the difficulty for the administrator to manage the monitored objects. Once the number of monitored objects reaches a certain level, the display of the list will look confusing and difficult to manage

Relatively independent monitoring methods will increase the difficulty of troubleshooting. If there is no correlation, you can only troubleshoot faults one by one along the monitoring objects that generate abnormalities. You cannot make simple and intuitive judgments on faults, which will lead to time spent on locating problems and solving faults. will be long

In addition, the alarm management function is not perfect, and the remote diagnosis and management of services cannot be realized.

For the above-mentioned problems in the prior art, still lack effective solution at present

Images