Terminal-cloud cooperative computing architecture, task scheduling device and task scheduling method

Abstract

The invention discloses a terminal-cloud cooperative computing architecture, a task scheduling device and a task scheduling method. The terminal-cloud cooperative computing architecture under an edge cloud computing platform and the task scheduling method under the architecture provided in the invention aim to solve the problems that the load of the existing edge cloud platform is high and the communication overhead when a task is submitted to a remote end is high in the rapid development cloud application time. Different from the traditional cloud computing platform, the terminal-cloud cooperative computing architecture disclosed by the invention is as follows: terminals, which are safely certificated and accessed to the system and have a certain computing capability, are also added into a computing resource pool, so that a terminal cluster is formed; the terminal cluster has relatively high expansibility; the cloud expansibility is increased; the terminal cluster and a service node cluster in static cloud complete a previous task computed by a single cloud platform together; the traditional cloud platform is optimized; the load of the cloud side is effectively reduced; the architecture has the real-time and rapid characteristics of edge cloud; and furthermore, relatively high and steady services of large data applications, such as the internet of things, can be provided.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a device-cloud collaborative computing architecture and a task scheduling device and method. Background technique [0002] The cloud era has enabled the rapid development of various services, and various intelligent applications have also followed. As many industries are moving towards digitalization, storage and computing are increasingly dependent on the cloud, and the cloud has become the hub of building an information society. In the era of big data, if all data is processed uniformly by the cloud data center, not only will it add too much load to the cloud, but also there will be a certain amount of communication overhead for submitting tasks to the remote cloud center. The traditional cloud platform will face many problems. Therefore, A data-centric edge cloud is produced. In the era of fast-growing cloud applications, the load of the edge cloud computing platfo...

AI Technical Summary

Problems solved by technology

At this time, due to the heterogeneity of resources, the differences in computing capabilities of terminal resources and clouds, the complexity of task scheduling is greatly increased.

[0005] At present, the research on the task scheduling algorithm of cloud computing is more based on grid computing, and adapts to the cloud computing platform by improving the traditional classic algorithm. However, due to the heterogeneity of resources, the computing power of terminal resources and cloud It is not suitable for the end-cloud collaborative computing under the edge cloud computing platform

Images