IMPROVED QUANTUM ANT COLONY ALGORITHM-BASED SPARK PLATFORM TASK SCHEDULING METHOD

摘要

A method for scheduling a task of a Spark platform based on an improved quantum ant-colony algorithm, which is related to the technical field of resource allocation and task scheduling. The method includes: collecting, by a node-information monitoring module, information of a state, such as central processing unit (CPU) utilization, memory utilization, and network transmission speed, of a node at which the non-edge-computing application is located; sending the information periodically to a node-information collecting module; updating, by the node-information collecting module, an indicator of the state collected by itself; sending, by a Spark scheduler, information of a to-be-scheduled task and a list of computing resource nodes that are currently available, to a task-schedule managing module; acquiring, by the task-schedule managing module from the node-information collecting module, information of a state of the computing resource nodes on the list; determining a scheduling scheme through an improved quantum ant-colony algorithm built in the task-schedule managing module; returning the scheduling scheme to the Spark scheduler; implementing the scheduling by the Spark scheduler. Duration for executing tasks can be effectively shorted, and efficiency for executing tasks can be effectively improved.