Container cloud-oriented task preemption scheduling method and system

Abstract

The invention discloses a container cloud-oriented task preemption scheduling method and system. The task preemption scheduling method comprises the steps: storing tasks submitted by a user through atask queue; taking out to-be-scheduled tasks from the task queue; judging whether the resources in the container cloud platform can meet the execution of the to-be-scheduled task or not; if the execution of the to-be-scheduled task can be satisfied, scheduling the to-be-scheduled task to the resource node, otherwise, destroying the low-priority task in the container cloud platform to release the computing resource, and scheduling the to-be-scheduled task to the resource node when the resource in the container cloud platform can satisfy the execution of the to-be-scheduled task. According to the task scheduling method and device, the tasks with low priorities can be destroyed and computing resources can be released in a resource full-load operation scene, so the tasks with high priorities are executed in time, meanwhile, due to smooth execution of the tasks, the number of the tasks waiting for being executed in the task queue is greatly reduced, and the task scheduling efficiency is effectively improved.

Description

technical field [0001] The invention relates to a resource scheduling management technology of a container cloud platform, in particular to a container cloud-oriented task preemption scheduling method and system. Background technique [0002] In recent years, container technology and related applications have received more and more attention at home and abroad, and the research and development and application promotion are developing rapidly. In foreign countries, container technology has formed a relatively mature ecosystem; in China, financial companies, Internet companies, and IT companies are actively investing in container technology research and development and application promotion, and the development momentum is rapid. At present, container cloud orchestration and scheduling based on container technology, automatic elastic scaling, QoS guarantee, security and privacy protection are long-term research priorities and difficulties in the field of container cloud. Amon...

AI Technical Summary

Problems solved by technology

[0004] In the traditional task scheduling environment of the container cloud, for scenarios with sufficient resources, this task scheduling mode can meet business needs, and any resource-demanding business can be scheduled normally. In a very demanding application scenario, a high-priority task is run when the overall resource usage is full. Since the resources cannot meet the business needs, the high-priority task will always be in the "waiting" state and cannot get a timely response. It is unacceptable for real-time high-priority tasks, especially in the field of national defense and security. , the untimely execution of high-priority tasks may lead to devastating disasters. In addition, when the "waiting" container tasks continue to increase, it will also affect the scheduling speed of new container tasks, thereby affecting the overall container cloud platform scheduling performance.

[0005] Existing container cloud-oriented task scheduling methods more or less have the following defects: (1) Tasks cannot be scheduled

When the cluster is running at full capacity, high-priority tasks cannot be scheduled, which will cause high-priority tasks to not be executed in time, which is unacceptable for real-time high-priority tasks

(2) Resources cannot be released: Since the traditional task scheduling mode does not preempt scheduling, when the cluster is fully loaded, if you need to schedule high-priority tasks, you need to wait for the low-priority tasks to finish running and then automatically release resources or manually release low-priority tasks. High-level task resources, unable to realize fast and automatic release of low-priority task resources

(3) Affect the new task scheduling speed

In a large-scale cluster, when there are many "waiting" tasks, these waiting tasks will repeatedly rejoin the task queue and reschedule after the scheduling failure, which will have a great impact on the scheduling speed of new tasks

Images