Load task migration algorithm for power optimization in marginal computing environment

Abstract

The invention discloses a load task migration algorithm for power optimization in the marginal computing environment. The algorithm includes: monitoring in a marginal computing cluster to acquire CPU (central processing unit) utilization of nodes and temperature of the nodes; comparing the CPU utilization of the nodes and the temperature with preset thresholds respectively and determining all superload nodes; performing random arrangement on all the superload nodes to generate a superload node queue and generating a migration target queue according to migration targets; making a statistic of to-be-migrated tasks according to the minimum-number migration task principle and generating a to-be-migrated task queue; selecting the nodes with the shortest migration time in the new migration task queue and taking the nodes as the migration target nodes of the to-be-migrated tasks; generating a migration list according to corresponding relation of the to-be-migrated tasks and the selected target nodes to generate a task migration scheme. With the algorithm, violation and performance reduction in a service level agreement of the marginal computing system can be reduced, and task execution is enabled to be more equilibrium and efficient by equalizing power consumption of the marginal computing system.

Description

technical field [0001] The invention relates to a power consumption-optimized load task migration algorithm in an edge computing environment, and belongs to the technical field of combining cloud computing technology with the application of Internet of Things technology. Background technique [0002] Edge computing is a service method based on Internet sharing of computing resources, storage resources, data resources and application resources, providing optimized computing services for other devices in a virtual computing environment. Edge computing is different from traditional cloud computing. Edge computing has an "edge" layer near the terminal. The "edge" layer is composed of various computing systems with different performance and more dispersed. Its basic structural framework is as follows: figure 1 As shown, with this "edge" layer, some cloud computing tasks can be performed at the "edge" layer without the need for Tasks are sent to the cloud computing center, becaus...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is to overcome the deficiencies of the existing technology, provide a load task migration algorithm based on power consumption optimization with the minimum migration time in the edge computing environment, and solve the problem that various nodes in the edge environment cannot be combined in the existing technology Node power consumption, CPU utilization and temperature, migration time and other metrics, and the problem of balanced scheduling of node tasks to reduce service level agreement SLA violations and performance degradation in edge computing systems

Images