Fault-tolerant strategy selection method based on markov Chain in fog environment

Abstract

The present invention is applicable to the technical field of fog computing, and provides a Markov Chain-based fault-tolerant strategy selection method in a fog environment. The method specifically includes the following steps: S1. For a fog system with N fog nodes, predict that there are i in the fog system. The steady-state probability π of a fog node failure i , i takes values ​​0, 1, 2,...N in turn; S2, detect π i Whether it is greater than the preset value, if the detection result is yes, the fault-tolerant strategy with the lowest comprehensive cost is output. The comprehensive cost is evaluated based on the price and time of the fault-tolerant strategy. Based on the steady-state probability of the fog system, a low-cost fault-tolerant strategy is selected, taking into account the time and cost of the fault-tolerant strategy, and improving the reliability of fog computing.

Description

technical field [0001] The invention belongs to the field of machine learning and distributed computing, and provides a Markov Chain-based fault-tolerant strategy selection method in a fog environment. Background technique [0002] In recent years, as people's tolerance for service delays has been decreasing, Cisco (Cisco) proposed fog computing technology in 2011 to reduce instant service delays. As a new computing mode, fog computing has rapidly become a research hotspot in academia and industry due to its advantages of low latency and high scalability. Among them, how to formulate fault-tolerant strategies to improve the reliability of fog computing technology has become an important research topic for many scholars. Existing fog computing technologies mainly serve new latency-sensitive network applications such as the Internet of Things, smart transportation, smart grid, virtual / augmented reality, and unmanned driving. Among them, fog nodes are mainly deployed near sen...

AI Technical Summary

Problems solved by technology

The second modeling tool based on Petri Net modeling has the limitation that the model is easy to become very large and cannot reflect the content of time

The above two fault-tolerant strategies based on preprocessing have a certain effect on promoting service reliability. However, because the real-time tasks processed by fog nodes are often simpler than other distributed systems such as cloud nodes, and their storage, computing and communication capabilities are lower than those of cloud nodes. The nodes are weak, and important data and processes need to be saved to the cloud. Multiple backups of task data and processes will cause waste of storage space and bandwidth resources. At the same time, the fog system also has the problem of complex state changes and dynamic changes in reliability

Images