A task allocation method in an industrial distributed data acquisition system

Abstract

The present invention proposes a task allocation method in an industrial distributed data collection system. This method comprehensively considers multiple factors such as collection node resource utilization, collection efficiency, load balancing, collection reliability, heterogeneous collection nodes, etc., and establishes an initial simulation-based The optimization model for the number of collection nodes and the optimization model for initial task allocation optimize the number of working collection nodes based on the actual configuration and usage of each collection node, improve the utilization of collection resources, and realize various tasks in the industrial distributed data collection system. Under different conditions of collection node resource configuration and resource usage, the initial assignment of tasks includes both 1:1 redundant tasks and non-redundant tasks, which meets the real-time and reliability requirements of industrial distributed data collection in the industrial big data environment. , the need for effective utilization of resources, etc.

Description

technical field [0001] The invention belongs to the technical field of data acquisition, and in particular relates to a task allocation method in an industrial distributed data acquisition system. Background technique [0002] With the advent of the industrial big data environment, the data sources in the industrial process are increasingly diversified and the data scale is increasing. Facing the collection of large-scale high-frequency data in the industry and some new application requirements, in order to ensure the sequential and real-time data collection In terms of reliability and reliability, more and more enterprises are beginning to consider adopting distributed systems for industrial data collection. In the design process of the distributed data acquisition system, the task allocation strategy is a very critical link, which will directly affect the resource utilization efficiency and data acquisition efficiency of the acquisition system. [0003] In a distributed e...

AI Technical Summary

Problems solved by technology

Although the above-mentioned patent can realize the effective allocation of tasks, it is only aimed at the situation where the resource usage of each collection node is the same at the initial allocation time. For heterogeneous collection nodes with different resource configuration or usage, the above allocation scheme is not applicable. Moreover, the allocation method described in the existing patent is only for the occasions where all tasks are important tasks and need to be backed up, or all tasks are not backed up. The acquisition reliability of the acquisition task is often only backed up for some important data in the acquisition task, and other tasks are not backed up. The task allocation method in the existing patent cannot achieve load balancing of each acquisition node in this case. Moreover, the existing patents do not provide a specific optimization method for the number of initial work collection nodes, which is likely to cause waste of collection resources

Images